User Psychology: Re-assessing the Boundaries of a Discipline

doi:10.4236/psych.2010.15041

Paper Menu >>

Journal Menu >>

Psychology

2010. Vol.1, No.5, 317-328

User Psychology: Re-assessing the Boundaries of a Discipline

Pertti Saariluoma1, Antti Oulasvirta2

1Cognitive Science, University of Jyväskylä, Jyväskylä, Finland;

2Helsinki Institute for Information Technology HIIT,

Aalto University and University of Helsinki, Helsinki, Finland.

Email: ps@jyu.fi

Received June 3rd, 2010; revised June 18th, 2010; accepted June 21st, 2010.

Currently, efforts of psychologists to improve interactive technology have fragmented and the systemization of

scientific knowledge stalled. There is no home for integrative psychological research on computer use. In this

programmatic paper, we reassess three meta-scientific issues defining this discipline. As the first step, we pro-

pose to extend the subject of study from the analysis of human mind in the interaction to the broader view of

human as an intentional user of interactive technology. Hence, the discipline is most aptly called user psycholo-

gy. Secondly, problem-solving epistemology is advocated as an alternative to the notion from natural sciences

that progress in science involves increased truth likeness of theories. We hold that implications to design is only

one outcome of psychological work—user psychologists should help solving the problems of other stakeholders

of technology as well. Thirdly, to help integrating fragments of research, analyses should be re-organized around

explanatory frameworks that can span multiple technical application areas. An explanatory framework combines

a problem domain, relevant knowledge, and the logic of scientific inference. To conclude, we argue that tech-

nology has become so pervasive an aspect of modern life that its relationship with human mind deserves the sta-

tus of a basic research question and its own discipline. Psychology of the computer user should not be the

handmaiden of technologists but define itself by its own terms.

Keywords: Psychology HTI User

Introduction

User psychology denotes the investigation of human use of

technology by means of psychological methods, concepts, and

theories. User psychologists study human action and its psy-

chological preconditions in the setting of technology use, with

an aim to contribute to pragmatic efforts in improving technol-

ogy. Central in user psychological work is identification and

analysis of use-related phenomena in terms of psychological

theories. Deep understanding of the mental life of computer

users can be useful for many stakeholders of interactive tech-

nology, including design, engineering, legislation, education,

and decision-maki ng , among other areas. In constructive pro-

jects, user psychology can help transforming intuition-based

practices to evidence-based practices and thereby improve

technology. This paper is about user psychology—what it has

been and, more importantly, what it should be.

The goal of the paper is to solve an important dilemma char-

acterizing the status quo. On the one hand the mode of psycho-

logical inquiry laid down for us in the eighties has been criti-

cized numerous times as too delimited and incompatible. On the

other hand, there nevertheless are active and seemingly fruitful

areas of psychological inquiry in human-computer interaction

(HCI), but these areas apparently do not fit in any existing me-

ta-scientific characterization of a discipline that includes psy-

chology. We believe it is the time to revisit the foundations of

the discipline and to reconsider what it entails. A careful ap-

proach would be to update the meta-science only minimally, to

the extent that these divergent areas could be subsumed. A

more future-oriented way would be to update the meta-science

so that its capacity to generate new perspectives and organize

research was increased.

The basic idea of searching for and exploiting psychological

knowledge in the development of technology has existed for a

long time and exhibited different forms. Before systematic lines

of inquiry can be said to have started, many pioneers were in-

spired by psychological concepts: Alan Turing and Douglas

Engelbart by the concept of intelligence, J.R. Licklider by

problem-solving, and Vannevar Bush by memory. Much of the

early human factors research was also psychologically oriented

(Welford, 1968). The end of the seventies and the beginning of

the eighties saw systematic lines of inquiry being established to

examine the use of computers. Researchers (James, 1973;

Sackman, 1970; Schackel, 1959; Shneiderman, 1976; Weinberg,

1971) began to scientifically investigate human cognition in

computer use and especially in the domain of programming.

These studies were closely connected with the research in skills

and expertise (Chase & Simon, 1973; de Groot, 1965, 1966).

One illustrative finding was that programmers become better

programmers by becoming better in remembering and “chunk-

ing” the parts of code that are not visible on the display

(McKeithen, Reitman, Rueter, & Hirtle, 1981).

Although not the first to discuss the use of psychological

knowledge in the study of computer use, Moran (1981) was the

first to propose a wider description than behaviorally and cog-

nitively oriented human factors and ergonomics. The term user

psychology was suggested by Moran (1981) in an introduction

to a special issue of ACM Computing Surveys entitled The

Psychology of the Computer User. He called that special issue

“the debut of user psychology.” The selling point was that de-

sign processes would be based on solid psychological prince-

ples instead of “folk psychology” (Moran, 1981). However,

P. SAARILUOMA & A. OUL ASVIR TA

318

Moran’s original formulation has been cited relatively seldom.

It has not been systematically elaborated afterwards and has

remained something of a curiosity. Moreover, in the subsequent

book The Psychology of Human Computer Interaction (Card,

Moran & Newell, 1983) the concept of user psychology was

considerably narrowed (Clemmensen, 2006)—perhaps because

the book presented the first model of the computer user, pro-

posing a prediction-oriented simulative approach as the ideal of

the new discipline. Although Card et al. did not rule out other

sciences, their focus on cognition was found too limited a view

to interaction. More controversial was their endorsement of a

“hard science” approach, proclaiming that quantitative predict-

tion should be the goal of research (Carroll & Campbell, 1986;

Newell & Card, 1986).

As we know, not everyone felt satisfied with the original

cognitive scientific description of the discipline, and the eighties

and the nineties witnessed several “divorces” from the main-

stream. Alternative agendas started to emerge that either de-

fined them in terms of how they differ from the “cognitivist”

orientation or were interpreted that way by others (Bødker,

1989; Carroll & Campbell, 1986; Carroll, 1991; Ellis & Nutt,

1980; Grudin, 1990; Heath & Luff, 2000; Helander, 1997;

Kuutti, 1996; McCarthy & Wright, 2004; Picar d, 1997; Such-

man, 1987; Winograd & Flores, 1986).

Psychology of programming was deemed problematic due to

its inability to derive applicable and specific enough knowledge

for developers (Carroll 1991, 1997), and during the decade that

followed, several important applications of psychological

science emerged to address needs of practitioners: user-cen-

tered design (Norman & Draper, 1986), cognitive task analysis

(Annett & Duncan 1967, Diaper & Stanton 2004, Kirwan &

Ainsworth, 1992; Schraagen & Chipman, 2000), ethnography

in human factors research (Blomberg, Burrell & Guest 2002)

and scenario-based design (Carroll, 2000; Rosson & Caroll,

2002). Testing methods were derived for the purposes of de-

signers, engineers, and information systems scientists (Nielsen,

1993). Usability engineers were perhaps less able to exploit

psychological knowledge with the same rigor as the cognitive

modeling camp and turned out to become a testing-centered

practice seeking standardized procedures. These applications of

psychology to both design and engineering have become

somewhat detached from contemporary psychology; main-

stream psychology is referred to quite rarely.

If the first two divorces were motivated by pragmatic needs

(of designers and engineers) and theoretical disputes, the third

“divorce” has been due to the landslide development of int era c-

tive technologies during the past two decades. Or instead of

divorce, one could describe this development as “fragmentation

across topics, theories, methods, and people” (Olson & Olson,

2003). Good examples of recent areas that are “psychologically

loaded” include ubiquitous computing (Weiser, 1991), affective

computing (Picard, 1997), ambient displays and tangible inter-

action (Ishii & Ullmer, 1997), attentive interfaces (Vertegaal,

2003), and mixed-initiative interfaces (Horvitz, 1999). Many of

these and other emerging areas clearly touch psychological

themes and needed to entice psychologists. However, research

has been technologically rather than psychologically defined,

and most work has been rather experimental. These circum-

stances are unfortunate, because they have led to a myopia

which has hindered grounding the work to relevant bodies of

knowledge in mother disciplines and countered the accumula-

tion of findings and elaboration of theories.

We believe that enough criticism has been voiced and it is

time to face the facts and address the problem. It is time to turn

attention to the most elementary demarcations that define a

discipline. The problems of the above type cannot be solved by

conducting new studies; for this, foundational work is re-

quired—the analysis and critique of suppositions behind the

paradigms (Saariluoma, 1997). To be constructive, we need to

propose several “updates” to the meta-science of this discipline.

Ultimately, in order to avoid being a slave to engineering, psy-

chologists must start defining their work according to psycho-

logical themes rather than in respect to technological bounda-

ries. This means re-thinking how we answer such questions as

“what is the subject of psychological research in this area”,

“what are psychological theories about”, “what kind of theory

is useful,” and “what is the role of psychological research in

constructive efforts like design.” In this paper, we address these

question by considering the broadest possible terms of any

discipline: subject of inquiry (the user in action), the purpose of

scientific work (problem solving), and the mode of scientific

inference (explanation).

Our agenda is threefold. Firstly, we propose to extend the

subject of study from the interaction loop to the human as an

intentional actor, human in the role of using a computer for

something. Hence, we propose readopting the term user psy-

chology. From this perspective, user psychology becomes in

part analogous to other applied psychology disciplines that also

study the human in specific activities, such as traffic psycholo-

gy, aviation psychology, counseling psychology, or political

psychology. However, it is evident that some revisions to the

previous notions of user psychology must be introduced. We

believe that areas defined around the terms “interaction” or

“cognition” or “factors” are too narrow. It is important to em-

brace a deeper view of humans as actors rather than as factors

(Bannon, 1991). Although we propose redefining the subject of

inquiry, we refrain from proposing an ideal method, as history

has shown such attempts to be problematic and restricting

progress.

Secondly, we propose a problem-solving epistemology (Lau-

dan, 1977; Saariluo ma, 1997, 2005), the core idea of which is

that progress in science is tantamount to increased ability to

solve problems (Laudan, 1977). In the setting of user psychol-

ogy, these problems can be empirical, conceptual, or construc-

tive by nature. Here, scientific advancement is marked not by

getting closer to “truth” but by increased problem-solving abil-

ity. This epistemology is different from the traditional episte-

mology coming from natural sciences, but it fits excellently to

the practical nature of contemporary thinking manifested well

for example in the CHI conference’s (CHI Conference, 2009)

requirement for authors to explicate their paper’s “contribution”

(to the body of knowledge) and “benefit” (for practitioners).

The problem-oriented view also points out that results of user

psychology should not serve exclusively designers and engi-

neers, but any and all stakeholders of interactive technology

from citizens to customers to decision-makers. We echo many

before us in the ideal of applicability of work; it is important to

take the pragmatic origins and ambitions of HCI seriously.

Despite the apparent lack of systematic theories, HCI as a dis-

cipline has been a moderate success both in the industry and in

P. SAARILUOMA & A. OUL ASVIR TA

319

the academia, but it could and should do even better. User psy-

chology should not get “off the hook” as happened with soft-

ware psychology (Carroll, 1997). But neither is we proposing a

“light weight” version of psychology for practitioners to exploit,

as that has been problematic as well.

Our third proposition is put forward in order to help reor ga-

nizing fragments of research currently splintered across tech-

nology-defined boundaries. We believe that good research

should not be like a black box that reports results from mea-

surements, but rather entail an in-built interest for systematic

development of understanding. The way to shed light on recur-

ring practical problems is to explain things, not only describe or

measure them. This way, user psychology could help research

already at the outset to set its goals more ambitiously and the-

reby overcome the problem of locality of results. The essence

of our proposition is to build a bridge between scientific prob-

lems and useful solutions via psychological knowledge. What

we call an explanatory framework combines a problem, rele-

vant knowledge, and the logic of scientific inference. How

these three components go together determine the applicability,

specificity, and therefore the problem-solving capacity of a

theory. The goal of restructuring an area around a general ex-

planatory framework is systemization of psychological know-

ledge in this area and for other areas that are analogous. Impor-

tantly, we do not propose that the boundaries of these frame-

works should be defined by boundaries within general psy-

chology. Instead, user psychological areas should be defined by

boundaries inherent in the phenomenon we are addressing:

human use of interactive computing. In the history and present

of HCI, many good examples of explanatory frameworks can

be found, from mental models to motor control. However, that

kind of integrative “basic research” has almost stalled during

the last decade. Our proposal is that we raise explanation to a

center stage in the new user psychology.

The main ambition of the paper is to explicate and re-assess

the boundaries that define our discipline, with the purpose of

re-directing psychologists to contribute to the broader bodies of

knowledge that surpass individual application domains. We are

proposing this programmatic agenda not for engineers or de-

signers but for psychologists already working with technology-

related issues. Researchers trained in psychology and already

working with technology have both the theoretical competence

and substance competence to formulate research questions in a

manner that the solutions sought for them would have practical

importance. The user psychology we are proposing will thus be

an applied discipline from two perspectives: Firstly, because it

is in intellectual debt to psychology and, secondly, because its

main purpose is not search for truth but problem-solving. How-

ever, technology has become so pervasive an aspect of modern

life that its relationship with the human mind is regarded now

as a key conceptual problem of the information age. The prob-

lem of user psychology therefore deserves the status of a basic

research question within psychology. We see it necessary that

psychologists start taking computer use more seriously and

eventually help developers to transcend testing-based practices

and address the issue in psychology’s own terms and rigor.

User—The Subject of Inquiry

The motif of a human agent working with a technological ar-

tifact is at the center of disciplines involved with the use of

technology, such as CSCW and Workplace Studies, CMC, HCI,

Human Factors, Ergonomics, and Information Systems. It is

partly due to conceptualizations of the user that the preceding

versions of user psychology have been found problematic and

inadequate. Some approaches have not been necessarily contra-

dictory or complementary, but have made tacit assumptions that

have limited generality. Therefore, as banal as it may sound, the

term user deserves attention. This, after all, is the subject of

inquiry in use r psychology. For proposing a synthesis, let us

examine and learn from three major conceptions of the user in

con temporary research.

First, the cognitivist school of HCI grounded their analytical

work to Alan Turing’s model of human mind that was later

expanded by Newell and Simon (1972) into a model of human

mentality. The point was to investigate users as information

processing systems characterized by successive information

manipulation operations (Anderson, 2002; Card, Moran, &

Newell, 1983; Laird, Newell, & Rosenbloom, 1987; Meyer &

Kieras, 1997; Gray, 2007). The most popular implementations

have been based on production systems where information

processing is a series of situationally triggered information

manipulations guided by rules of the form: if P (attern) then

Action). In spite of the undeniable contributions of these models,

some have remarked that they have little to say about emotional

interaction, experience, motives, meaning, personality, or cul-

ture. Another counter-argument is that these models are mecha-

nistic in their assumption that mental processes can be reduced

to causal sequences (Dreyfus, 1992). These are not newly dis-

covered limitations but were familiar already in the 1930s to

Wittgenstein (1958) in his critical analysis of the Turing ma-

chine. For user psychology, the main problem is however

whether a causalist computational language really has sufficient

power of explaining intentional action (Searle, 1983). However,

although there are counterarguments and limitations in the

modeling approach, the core point of these models remain true:

that internal mental events and properties of human mind can

explain (some) observed behavior of the user.

Second, early activity theorists in HCI (Bannon & Bødker,

1991; Bødker, 1989; Kapteli nin, 1996; Kuutti, 1996; Nardi,

1996), whose rhetoric in many cases built on claimed inadequ-

acies of the cognitive theories, drew from the Marxist pragmat-

ic philosophy. The core idea of Marxist philosophy was that

humans are what they do, praxi s shapes them and is itself

shaped by the institutions within which they function

(Bernstein, 1971). Leontyev (1981), one of the founding fathers

of activity theory, writes: “the human psyche’s properties are

determined by man’s real relations with the world, which de-

pend on the objective, historical conditions of his life” (p. 268).

In present-day activity theorists’ interpretation of interaction,

the main unit of analysis is activity (Kuutti, 1996). Activities

are driven by needs—not necessarily an individual’s but an

organiz ation’s—where a certain goal state is to be achieved.

Activities themselves consist of actions of individuals that can

be mediated by instruments or tools. Ergo, the core of being a

user means utilizing a technological artifact in an activity in

pursuit of a goal and satisfaction of a need. Technology is in a

mediating role in human pursuits (Kuutti & Kaptelinin, 1997).

While this formulation is acceptable, the most evident problem

in the interpretations of contemporary activity theorists in HCI

P. SAARILUOMA & A. OUL ASVIR TA

320

lies in their reluctance to systematically include mental aspects

in their explanations of human action, although both Vygotsky

(1980) and Leontyev (1981) were keenly interested in brain

functions, consciousness, and cognitive capabilities (Roth &

Lee, 2007). This stance closes out of discussion much of the

psychological work realized and advancements achieved in

psychology after the first half of the 20th century. The intuitive

externalism—that is, the tendency to emphasize external di-

mensions of human action in constructing theory languages—

has lead activity theorists to emphasize collaborative aspects of

interaction, which of course is fruitful, but has largely disre-

garded known social psychological work on the same topics.

Because of this, as activity theory is presently explicated, it

does not provide a sufficiently elaborated model of user for

modern user psychology. Its connection to modern empirical

tradition of psychology remains shallow and it does not en-

courage efforts to integrate findings across disciplines. One

consequence of intuitive externalism is, as Nardi (1996) ob-

served, that the tradition is more descriptive than explanatory.

Third, phenomenology (Grossman, 1984) has provided an

alternative, first-person view, to the study of user. In pheno-

menological accounts of human-computer interaction, the cen-

tral feature of human action and experience is intentionali-

ty—being directed towards something. The philosophical roots

of phenomenology involve authors such as Husserl (1913/1982),

Heidegger (1927/1962), Gadamer (1960), Merleau-Ponty (1962)

and Schütz (1932). Encyclopedia of Phenomenology (1997)

distinguishes seven types of phenomenology, of which the most

cited type within HCI has been Martin Heidegger’s (1927，1962).

Heidegger held that our activities are always “in the world” and

we must not study these activities by bracketing the world, but

by looking to the contextual relations that our activities and

meanings have in the world. However, the concept of world

does never in phenomenology refer to the naïve material world

but it entails all our experiences and meanings. Phenomenolo-

gists take meaning and experiences of meanings central in their

analyses of human-computer interaction. Dourish (2001) has

opened up these notions for HCI by the terms intentionality,

ontology, and intersubjectivity. Intentionality means the direct-

ness of actions and the intended effects of those actions, ontol-

ogy concerns the ways in which we come to understand the

computational world, and intersubjectivity reflects sharing of

the world with other individuals. Importantly, these are not

creations of the mind, but achieved in interaction with the ma-

terial, social, cultural, and historical conditions of the world.

Being a user boils down to one’s experiences and meanings of

“being in the world” achieved by using one’s body to interact

through technological artifacts. This characterization highlights

the constructive relationship between the user’s felt experience

and intentions on the one hand and the material-soci al-cul-

t ural-historical conditions on the other. This approach brings

about many sophisticated conceptual distinctions which are

more or less absent both in cognitive and activity research. We

agree with the claim that interaction is to be understood in

terms of how people act through technology, not on it (Dourish

2001). Similarly to Winograd and Flores (1986), Dourish (2001)

expands, for example, on Heidegger’s idea of a tool being ei-

ther “ready-to-hand” or “present-at -hand.” The experience of

objects depends on our actions and intentions—one and the

same computer is seen as a different object by programmers,

information systems scientists and secretaries, because their

stances differ as a consequence of their work. If one admits that

psychology is the study of mental processes and human beha-

vior, there is no a priori reason to rule out phenomenological

analyses from interaction discussions. Simulative cognitive

science does not really capture experience by its concepts and

consequently it remains silent on this phenomenon. Activity

theorists do not refer to subjective or internalist conceptualiza-

tions and therefore they also have had little to say about felt

experience. Phenomenologists are right in that experience exists

and they have done an important service to the community by

taking it under scrutiny. Methodologically and theoretically,

however, phenomenologically oriented work has remained far

from the main body of psychology. Phenomenologically

oriented researchers have done relatively little experimental

work; instead, they have made conceptual analyses and prof-

fered folk psychological considerations on interaction (Svanaes,

2001). There is, of course, no apriori obstacle for experimenta-

tion as mode of inquiry among others. Moreover, among many

other findings on experience, psychology has demonstrated that

experience is not an island but rises as a consequence of other

mental processes. Freud was already able to argue that con-

scious experience is generated by the unconscious. To under-

stand consciousness we have to understand unconscious

processes and include in our explanations emotional, social, and

cognitive processes also at the neural level.

In sum, all three traditions reviewed entail important but dif-

ferent contributive elements for understanding the user, and

they also have their limits. Knowledgeable of these, it is possi-

ble to begin to sketching a new way by taking contributive

elements and unifying them into a new picture of a conception

of a user. We propose viewing interaction first and foremost

from an intentionalist perspective, distinguishing intentionality

as the primary characteristic of the user. Interaction is action

and the user’s intentions are the source of personal significance

and the guiding principle of that action. Thus, with phenome-

nologists we share the idea that human intentions are vital in

explaining human actions. While material objects are subject to

laws of natural sciences, human actions are intentional and

goa l-driven. The locus of scientific explanations of human ac-

tion is in the future, what the human being intended to achieve

(von Wright, 1972). I use my cell phone to call my friends in

order to meet them and to socialize with them. Notions like

usability, errors, and task performance can only be understood

against this background. However, we should accompany phe-

nomenological theorists only this far.

In fact, we believe we must give up the first person perspec-

tive and, instead, we should look at intentions from the third-

person perspective, in order to save the objectivity of science

and possibility to penetrate into subconscious mechanism as

well as brain-based explanations. This perspective of looking

consciousness from the outside has traditionally been called

heterophenomenological (Dennett, 1991). In practice, it means

that we can apply such methods as interviews and protocol

analysis to study conscious experience (de Groot, 1965; Erics-

son & Simon 1984). Cognitive and other areas of psychology

have a lot to add to the notion of intentional agent. Let us as-

sume that a person with an intention to write LISP code forgets

something as the complexity of the required representation

increases (Anderson, Farrell, & Sauers, 1984). The explanation

P. SAARILUOMA & A. OUL ASVIR TA

321

of a failure of this kind is seldom in intention, but it must rather

be searched from operations and capacities of the contributing

psychological mechanisms, without which complicated actions,

such as finding a link on a web page, or finding a menu item,

are impossible.

In a sense, there is nothing new in the intentionalist concep-

tualization, as we habitually talk about users’ goals and needs

in HCI. Nonetheless, the nature of these guiding principles has

not received the importance as it should have. For example,

“user needs”—such as “need to drive screws fast, faster than by

hand,” in the context of screwdriver design (Ulrich & Eppinger,

2003)—do not often describe psychological needs at all but a

s ought-for performance of an artefact. Thus, the very nature of

intention in interaction should be subjected to deeper scrutiny.

Essentially, intrasubjective processes should contribute to ex-

planations of intentional action. The positive side of this take is

that the intentionalist view can help us to escape the criticism

that psychologists too easily view interaction mechanistically

and as a closed system, or through “boxologies” that reduce

personal meaning into operations in information processing.

We propose raising intentionality explicitly to the subject of

inquiry.

We believe that also in practical attempts at improving tech-

nology, intentional action is the natural phenomenon to be ana-

lyzed, understood, and designed for. The reason for this is log-

ical: All technology is designed to support people in their at-

tempts to reach their life goals, which are set in their actions.

We rely on computers when we drive cars, when we use mobile

phones or elevators, when communicating via computers or

when using Internet services. If one takes serious the idea of the

user as intentional actor, the conclusion is that user psychology

cannot be defined in terms of the technical quality of the com-

puter in question; that is, whether the computer is a computer in

the Turing sense. Any interactive computer even embedded

computers in cars and digital wrist watches that are computa-

tionally si mple are in the domain of user psychology. Looked at

from this perspective, the psychology of the human user should

be one of the key topics of the sciences of the artificial (Simon,

1969).

A Problem-Solving Epistemology

“In appraising the merits of theories, it is more important to

ask whether they constitute adequate solutions to significant

problems than it is to ask whether they are ‘true,’ ‘corroborated,’

‘well-confirmed’ or otherwise justifiable within the framework

of contemporary epistemology” (Laudan, 1977).

User psychology and related disciplines are in a way difficult

to locate on the map of science. On the one hand their ultimate

goal of improving technology for human use makes them ap-

plied science. In user psychology, psychology is used for solv-

ing practical problems, analogous to, say, what is done in traffic

psychology. Bunge (2001) defines a discipline as applied if it is

in intellectual debt to another field. On the other hand, we must

ask whether user psychology is really applied science, for its

goal is to produce a new coherent view of the human’s rela-

tionship with interactive technology. If this kind of problem

that addresses the most salient characteristic of our time is not

basic science then what has went wrong? In this light, the old

basic–applied dichotomy is problematic.

We have to look what science is in a new way. Instead of

seeing science as a set of justifiable beliefs, or improvements in

“truthlikeness” of theories, we might see it as a dynamic

process where our problem-solving ability is increased (Laudan,

1977). This means that scientific concepts, theories and empir-

ical findings should be evaluated on the ground of practical

advancement they allow us in developing technology or if for-

mulated in a wider sense: how they improve the development of

information society. Naturally, problem-solving capacity can-

not be increased on the basis of outright false theories, so the

two aspects of epistemology—truthlikeness and usefulness—

are related. However, we see that the primary goal of user psy-

chology should be not be truthfulness, as there are cautionary

examples in its past (Carroll, 1991, 1997).

What does a problem mean? In his work, Laudan (1977) dis-

tinguishes two kinds of problems: empirical and conceptual.

Empirical problems involve unknown phenomena, unknown

factors, and unknown effects. Much of user studies and evalua-

tive experiments fall into this category. Conceptual problems

involve implausibility (to explain something within a theory/

concept), inconsistency, and incompatibility. Most theory ap-

plication and construction addresses this category of problems.

Because Laudan was mainly interested in natural and humanis-

tic sciences, it may have escaped his attention that there are also

problems of constructive character. Three subtypes can be

thought of: 1) inability to imagine a solution, 2) insufficient

knowledge for implementing a solution, and 3) unavailability of

resources for implementing a solution. Much of user interface

design and UI software technologies work reported in academic

forums like UIST address these problems, but just as well deci-

s ion-making, policies, and educational programs could be

thought this way. The contribution that user psychology makes

toward solving these problems is that its theories and results of

empirical work can provide us with ideas on how to make good

constructions. The point is that work on all three types of prob-

lems is necessary. It is a form of short-sightedness to require

every study to be able to contribute to constructive problems

directly. In the absence of understanding, this form of prob-

lem-solving science would not carry far as it would lead to poor

solutions to the problems.

What is a good solution, then? Laudan (1977) argues that

progress in science is about adequate solutions to important

problems. Laudan (1977) defined a good solution as having

three qualities:

1. it solves a problem that is significant,

2. it has the potential to solve many new problems, and

3. it solves the problem(s) effectively.

From the perspective of the first quality, research on cogni-

tive models of the user has been successful, because these

models helped solving significant problems of the early 1980s.

However, one can say that in the 21st century its challenge has

been the second quality: it has not been able to solve those

problems that are the most important ones (Carroll, 1991), as

technology has developed and models were not updated accor-

dingly. Moreover, even the earlier versions of GOMS (Goals,

Operators, Methods, & Selection Rules; Card et al., 1983) were

criticized as being too difficult to apply—in other words, too

inefficient as a solution (3rd quality). However, significant

progress has been made in lowering the barrier to apply these

models to everyday interaction problems (John, 2004). Another

P. SAARILUOMA & A. OUL ASVIR TA

322

good example is Fitts’ (1954) law that has been applied widely

to analysis and prediction of input devices. The problem of

target acquisition is central to all HCI and Fitts’ law offers an

effective way to address it, which is why it fills two out of the

three qualities of Laudan. However, the law in its original form

does not really involve any explanation, rather a description of

statistical relationship between three variables (movement time,

target size, target distance). Without explanatory basis, its ex-

tension to new problems has mainly taken place through trial

and error rather than prediction (Friedlander et al., 1998).

Importantly, the problem-solving epistemology makes it

clear that prediction is only one aim in the psychology of the

computer user (Card, Moran, & Newell, 1983). From the Lau-

danesque perspective prediction is a false a goal for epistemol-

ogy. It is more essential to have psychological knowledge that

is applicable (Carroll, 1991). However, it is very common for

non-psychologists to believe that the existent psychological

knowledge is somehow applicable as such and ready to be used,

although it is often necessary to study first what the meaning of

a psychological construct in a particular phenomenon of tech-

nology use, before it can really be applied in practice. Thus,

identifying the most significant and practical explanatory frame-

works is as important task for research as is actually elaborating

them.

Thus, we believe that the core of our epistemology should

not be quantitative prediction of outcomes but explana tion. If

we think of a normal engineering design, it relies heavily on the

laws of nature (Pahl & Beitz, 1996). An engineer must be con-

fident that, once built, the bridge will stand. Neither do engi-

neers build skyscrapers by trial and error. The reason they can

do this is that they know the properties of materials and the

effect of forces on structures and can thus engage in explanato-

ry problem solving. However, when we move to interactive

technology, it is a completely different case: trial and error and

reliance on intuitions in place of theory-based argumentation.

This has been a necessity, because we have not been able to

create a practice in which the laws and principles of modern

psychology would support and guide problem-solving the way

natural physical laws do in machine engineering. Getting from

the intuitive to explanatory problem solving practices is a cen-

tral goal for user psychology. In other words, user psychology

is ideally conducted in a way where applicable solutions are

sought via deepening understanding.

But whose problems are we are attempting to solve? We be-

lieve that any piece of technology ties around it a complex web

of stakeholders whose problems are relevant and which we

ought to be solving. There is no a priori reason to limit our

work to serve designers. True, we need to aid designers to find

rational ways of solving immediate interaction design problems.

However, designers are not our only stakeholders (Carroll,

1991, 1997). There are also the individual end-users, or cus-

tomers, whose needs and abilities we have to worry about. By

informing consumer segments, user psychology can help them

making informed choices and raise awareness of the conse-

quences of poorly designed technology. But the list is much

longer. In organizations, there are stakeholders from adminis-

trators to helpdesk to managers to human resources specialists.

In the Internet, user communities and groups are formed that

have different requirements for technology. Developers also

have their constructive needs, as have designers, and service

providers and commercial companies have their take and inter-

ests that are often evaluated in monetary terms. Finally, institu-

tions like schools and societies as a whole take significant deci-

sions on information technology, and thus their needs should

also be served. Given these widening spheres of stakeholders,

we believe it has become impossible to maintain that design is

the main “client” of our discipline. The ultimate goal of user

psychological research should be the improvement of the con-

ditions of human life and our research should indeed be eva-

luated, in the end, from that point of view (Cockton, 2008; Sel-

len et al., 2009).

This characterization fits quite well the de facto practices of

the HCI community. The flagship conference CHI, for example,

places as the foremost demand that its papers “offer contribu-

tions that clearly and significantly advance the field of HCI.”

Moreover, the conference asks the authors to think “what kinds

of problems might readers be facing to which your contribution

could provide the solution” and tries “to make sure that the

submission explains the contribution in sufficient detail for the

full benefit to be extracted” (CHI Conference, 2009). In other

words, CHI has already started to define itself in terms of solu-

tions to stakeholders’ problems. Previously, however, we were

lacking an epistemology that would fit these aims. If user psy-

chology wants to contribute to real-world efforts, as it should,

this reformulation of its epistemology as a problem-solving

epistemology is natural.

Explanatory Frameworks

Prediction is the sine non qua of engineering sciences. When

engineers build a bridge or skyscrapers, they apply models

derived from the laws of nature to make advance calculations.

Although they also test their constructs empirically, the ad-

vance use of laws gives their work a predictive character. Of

course, there is always also an element of intuition involved.

The designers of the ship Estonia that sank in 1994 with 800

passengers did not have sufficiently sophisticated mathematical

formulas to calculate the weight of the wave forces against the

keel structures (The Joint Accident Investigation Commission

of Estonia, Finland & Sweden, 1997). Save the cognitive mod-

eling community, the predictive stance is weak or missing in

areas around human-technology engineering. Usability testing

is common, but it is one thing to generate and test and another

to predict and test. This is why much of present-day design

practices remind of medieval construction practices where, by

means of trial and error, stable solutions are found and passed

on from one to another.

Nevertheless, as we discussed in the previous section, we do

not suggest taking prediction as the ultimate goal of this discip-

line. There are reasons for this. First, it may be impossible.

Landauer (1997) criticized such attempts by claiming that the

to-be-predicted phenomena are too complex: Any attempt at

prediction is overly sensitive to differences in the starting pre-

sumptions, subject to extreme random variation, and includes

variables that are impossible to know in advance. Moreover,

people are not “deterministic” and thus never 100% predictable

(von Wright, 1972).

We believe that the primary epistemic goal of this discipline

should be explanation. By explanation it is generally meant that

connections are established between previously miscellaneous

P. SAARILUOMA & A. OUL ASVIR TA

323

items. However, explanation is not necessary for prediction－

statistical prediction involves no explanation. Instead, explana-

tion is often a sufficient condition for prediction, at least if it is

not conceived narrowly as quantitative prediction. Namely, the

nature of prediction can also be of a relative or nominal type:

“this design will induce user error,” or “design A will be better

than design B.” When connections are established between

variables our artifacts affect, such as number of items in a menu,

and variables of use, such as number of errors, an explanation

will have some predictive power.

How should we work toward explanation in concrete prob-

lems of computer use? In its simplest form user psychology

unifies 1) a problem with 2) the respective knowledge from a

field of psychology to chart solutions through 3) scientific in-

ference that maps the two. An explanatory framework is a psy-

chological knowledgebase which provides us with information

beneficial in solving a problem or some of its aspects (Saariluoma,

2005, Figure 1).

Problem * Scientific knowledge  Solution

Figure 1.

Explanatory framework. The problem is described in a

way that scientific knowledge can be brought to explain

it and a solution inferred.

In the rest of this section, we examine a handful of important

frameworks from the past and present lines of research: biolog-

ical psychology, information processing psychology, psychol-

ogy of emotions, mental contents, individual differences,

groups, and socio-cultural aspects. The list is only one possible

cross-section of vast volumes of work. The list is not supposed

to be taken as comprehensive, it only serves to illustrate two

points: 1) that there are many domains of general psychology

outside cognitive psychology that are in fact already being ap-

plied in contemporary HCI, 2) that the internal logics of the

frameworks are very different—so different that it is difficult to

see how any one framework could possibly cover them all. Any

concrete interaction problem is so complex that multiple ex-

planatory frameworks must be called in. Some of these frame-

works have emerged as a response to the needs of a particular

application area, but have widened their scope along with re-

search being formulated around user-related issues. For exam-

ple, the recent surge of research on emotions in human-computer

interaction was boosted by some seminal work on affective user

interfaces. By re-thinking psychological research from the

perspective of the user, user psychology may reach generality

across application domains.

It is important to note that the examples presented below

represent high or macro-level explanatory frameworks and

define rather extensive wholes. In practice, explanations at the

level of a particular study must often be much more nuanced

and refer to more detailed theoretical explanations. For example,

much of (but not all) information processing psychology that

has been applied to HCI has utilized what could be called a

ma cro-level explanatory framework of limited capacity. How-

ever, in a particular project one could be interested in micro-

level frameworks such as, say limitations of the human working

memory and its subsystems (Baddeley, 1986).

We begin with questions typical to a framework, and then

look at the background theories, and finally present concrete

examples of empirical work. It takes considerable time to de-

velop a good micro-level explanatory framework for a particu-

lar problem, but in many areas that work has been done to the

extent that we can start coalescing the micro-level explanations

into more coherent wholes. Looking at things from this pers-

pective forces us to address and improve the generalizability of

our work.

First, the limited capacities of human information processing

system is one of the classic explanatory frameworks and partic-

ularly strong in contemporary human factors (Wickens et al.,

2003). Human information processing is limited in many dif-

ferent ways; most well-known limitations relate to perceptual

discrimination (Goldstone, 1994), working memory capacity

(Baddeley, 1986; Cowa n, 2000), and selective attention (Pash-

ler, 1998; Styles, 1997). These limitations have been studied for

decades as explanations to performance and errors (Broadbent,

1958; Miller, 1951), and utilized in interface design to address

such questions as how to construct a display that is legible, how

to make warning signals discriminable, how many alternatives

you can have in a menu node, how to design notifications so

that they do not hamper memory, and how users search web

pages. A powerful example concerns perceptual discrimination.

People can visually perceive only a limited part of the electro-

magnetic spectrum from 380 nm to 750 (Coren, Ward, & Enns,

1999) and hearing is similarly limited, between 20 and 20 000

Hz. This means that sensation is based on a relatively narrow

spectrum of physical energy. Discrimination based on the elec-

tromagnetic spectrum may be difficult because people simply

cannot exceed the absolute thresholds of human sensory sys-

tems. However, discrimination depends also on the quality of

background, and if the background information is too similar

with the target, it also makes it difficult to perceive the object in

question. Low intensity of a stimulus may refer to a target re-

flecting light inadequately or to a target that is not clearly dis-

criminable against its background (Isler, Kirk, Bradford &

Parker, 1997). A somewhat different set of limitations is in-

volved in human attention (Pashler, 1998; Styles, 1997). Atten-

tion segregates the target or figure out of its background and

thus organizes perceptual reality in only one of the millions of

possible ways. Attention selects the messages, which are im-

portant for the ongoing action. In this way, it allows human

mind to focus on relevant things. The capacity of selection is

normally one unit at the time (Broadbe nt, 1958). It is possible

to switch attention from one target to another relatively swiftly

and thus follow two or even more competing messages at one

time. However, this kind of performance has its costs and is

risky. One of the core problems here is how we can select the

target information from irrelevant background information. A

key explanatory factor is discriminative cue. This means that

relevant information has some property such as color, which

makes it different from irrelevant information and thus is easy

to pick up. Discriminative cues and selective attention are im-

portant issues in graphical interface construction. It is critical to

be able to design the information to be expressed in such a way

that, from the action point of view, the most important informa-

tion is sufficiently discriminable. The system of discrimination

to be used is often a vital issue in designing effective visualiza-

tions (Card, 2003). These examples give an idea about how

capacity-based explaining is already part of user psychology. It

is necessary to ensure that interaction with an interface does not

P. SAARILUOMA & A. OUL ASVIR TA

324

surpass the available capacity. The main way of reacting is

naturally to reduce the complexity, which can be done either by

redesigning the interface in question or by improving the skills

of the user.

Another deeply seated framework concerns mental contents

of the user. Such notions as schemas and mental models have

often been used to describe information in mental representa-

tions of computing systems (Carroll, 1997; Norman, 1987).

What these notions share is that they involve beliefs or proposi-

tional information, properties of this information, and function-

al connections between content elements. The concepts of

mental contents provide us with an alternative perspective to

interaction compared to capacity-based explanations. The cru-

cial limit of capacity based research is that we can fill the li-

mited capacities with any type of information. That information

can be correct, false, or irrelevant but if it does not exceed the

processing capacity, there are no problems. As a metaphor, we

can fill a bottle with milk, honey, poison or any substance as

long as the bottle is not broken or does not overflow. This

means that the capacity based language cannot express those

problems which are characteristically explainable in terms of

mental contents (Saariluoma, 1997). One active research area

has been representations of objects of information that are not

directly present but “projected” on the user interface. The feat

of being cognizant or acting in the absence of instigating stimu-

li is of course not confined to human-computer interaction. For

instance, talking about a yesterday’s weather, counting time,

and thinking about our political system are such activities, and

in explanations of these phenomena we find links to general

psychology. There are many other interesting but less researche d

topics concerning mental contents related to computer use, such

as attitudes, values and social representations of users. Good

examples are the seminal studies of Reeves and Nass (1996)

t hat showed how users attribute humane qualities to computers

and media. Key in this explanatory framework are the mental

contents and inference processes that work on them that allow

the user to “jump into conclusions” beyond the literal meaning

of technological projections. Such inference is so basic to the

construction of meaning through interaction that the topic

should cover much more ground in human-computer interaction

than it currently does.

The third widely utilized mode of explanation concerns indi-

vidual differences. This macro-level framework subsumes mul-

tiple middle-level frameworks from skills to age to gender to

personality. The issue of individual differences has been recog-

nized since the very first experimental studies in HCI (Sackman,

1970) and is today reflected in popular agendas such as Uni-

versal Access and Design for All. Since we cannot possibly

cover all relevant frameworks, we briefly visit the issue of user

personality as an example. The best known of feature approaches

to personality is most likely the so-called BIG 5 theory which

focuses on five basic personality traits. These are extraversion,

agreeableness, conscientiousness, stability, and openness to

experience (McCrae & Costa, 1997; Wiggins, 1996). A typical

problem could deal with the question of how a user’s commit-

ment to a device might depend on a number of personality fea-

tures and the symbolical and identity values provided by the

product (Aaker, 1997). Another often used view to individual

differences is age (Hawthorn, 2000). Czaja and colleagues

(Czaja et al., 2006), for example, analyzed the connection of

intelligence and anxiety to older peoples’ use of computer

technology. Similar to all sub-frameworks is the idea that users

are unique and idiosyncratic characteristics must be taken into

account when analyzing interaction.

Fourth, a recent surge in the study of emotions in interaction

has opened up a very different system of explanatory grounds.

Emotions define our subjectively felt personal positions towards

events and objects (Jordan, 1998; Norm an, 2004; Power &

Dalgleish, 2007). Such categories of contemporary ICT re-

search as user acceptance and user experience are essentially

emotional issues. Of particular relevance among emotional

phenomena are, at least, the contents of emotions and their

connections to cognition and action. Emotions can be unders-

tood as reactions and preparations to prevailing situations, but

before we know what a situation is we have to make a cognitive

analysis (Power & Dalgleish, 1997). Emotions are activated

with appraisal, the cognitive analysis of the situation which in

user psychology involves a computer technology. Therefore,

the designers’ route to influencing user experience by means of

design is via cognitions (Beck, 1976). Emotions are also im-

portant, because they can convey information about our needs.

Naturally, these connections have an important role in motivat-

ing our immediate actions. Finally, emotions are always impor-

tant in experiencing the world in interaction with computers.

They bring some self-perspective to experience. They also tell

the “goodness” of the experience to the experiencing person.

We can also speak about a “felt” experience, which opens a

different perspective to interaction when compared to perfor-

mance (McCarthy & Wright, 2004). There are questions which

can be answered on emotional grounds only and for this reason

it makes sense to take emotions as an independent explanatory

framework.

Fifth, the recent “Decade of the brain” has brought about a

general movement toward biological psychology, also within

the HCI community (Parasuraman & Rizzo, 2007) that did not

recognize this area before the turn of the millenium (Landauer,

Helander, & Prabhu’s, 1997 handbook did not have a chapter

about it). The basic form of argumentation is familiar to us,

going from biologically determined properties of human per-

formance to explain why some design alternatives are prefera-

ble to others. Human motor performance is a prime example of

linking performance to properties of human nervous system

(Rizzo, Robinson & Neale, 2006). The size of hand and fingers,

muscle power, limb and joint movement, the speed of neural

signals, and physical limits of transduction are variables that are

crucial in designing low-level interaction. The harmony be-

tween movements and input devices is a critical issue and many

types of problems require signal processing analyses in a peri-

pheral and central level. Recently, neuroergonomics has ex-

tended the application of biological psychology from “low-

level” issues such as sensation, perception, and motor control to

problems related to higher-level cognitive concepts such as

vigilance and atte ntion and memory capacity. It has also been

applied to register physiological states of users, such as stress,

emotional states, and mental workload. Arguably, neuroergo-

nomics should be as important a part of user psychology as

neural theories are for understanding mind and it can provide

deeper understanding of user performance with pointing devic-

es than statistical models such as the Fitts’ law (Fitts, 1954). As

a philosophy of science, however, biological psychology may at

first blush appear too strict in its reductionism—a position that

has been criticized numerous times. It may seem that in many

P. SAARILUOMA & A. OUL ASVIR TA

325

important design solutions, there is no additional value that can

be achieved by biological explanations, but this is not always

true. There are indisputable cases in which neuroergonomics

can be of real and non-replaceable use in HCI. On the other

hand much of the research is still in a very early stage com-

pared to the cognitive paradigms.

Sixth, the development of information and communication

technologies—first groupware and now especially web and

recently mobile technologies—have opened new forms of com -

munication and collaboration between people. On an elementar y

level one may focus on the effects of group parameters such as

size when performing a basic interaction task such as searching

information from a display. Forlines and colleagues (2006), for

example, noticed that groups may be slower in searching in-

formation from a display but they are 10-15% more accurate

than individuals. On a higher level it is possible to investigate

motivational and dynamic factors for users interacting in groups

( B ackstrom et al., 2006; Beenen et al., 2004; Malone &

Crowst on, 1994). Sociometric analyses (Lindzey & Borgatta,

1954) and many other ways of studying groups have been taken

in use again. In general, group-level concepts can be used both

in analyzing interaction phenomena as well as in explaining the

forms they get. In a narrow sense, interaction means immediate

interaction with a screen and controls, but, in a wider sense, it is

essential to analyze also how groups of people organize their

actions when mediated by technology (Grudin, 1994).

Seventh, beyond the group level, cultural issues are on the

radar of user psychology as well. Humans are essentially cul-

tural creatures. There are numerous questions in computer use

that must be resolved by referring to culture, such as values,

communication practices, and cultural norms. We have to be

able to discriminate important subcultures and the ways they

see computer use, to be able to serve all segments of users.

Women, for example, form their own subcultural group in web

cultures (Frieze, Hazzan, Blum, & Dias, 2006). One additional

resource can be provided by cross-cultural psychology with its

long tradition of conceptual and empirical analyses of human

cultures (Cole, 1991; E kman, 2003; Hofstede, 1984; Matsumo-

to, 1996). There are much fewer obstacles now for communica-

tion between people of different countries and districts than

there have been during previous centuries. This means that

there are new ways of group formation in society and one can

participate with very different types of groups. Choi et al.

(2005), for example, reported a comparative crosscultural in-

vestigation concerning design norms for mobile data services.

In-depth interviews were conducted in Korea, Japan, and Finland,

to find a number of culturally explainable differences between

users. The most important of the differences were found in

uncertainty avoidance, individualism vs. collectivism, and con-

text, which could be linked with such usability features as va-

riety of options, contents, the use of colors, and iconic style.

Wider reference groups that are characterized by such attributes

as nationality, ethnicity, values, beliefs, living conditions, so-

cial order, or religious group will also become important in user

psychological analyses. Bourges-Waldegg and Scrivener (1998)

investigated multicultural CSCW and noticed that understand-

ing differences in representations forms a major design chal-

lenge. They very justly called attention to analysis of meanings

underlying cultural representations. One can also find numerous

subcultures within one’s own society. Age, work organization,

profession, wealth or education differences can culturally di-

vide people. Even a larger context for socio-cultural human-

technology interaction is proposed by Rauterberg (2006), who

sees socio-cultural computing as the new paradigm emerging

amongst the traditional personal and cooperative computing.

We have presented existent and emerging explanatory frame-

works referring to such subdisciplines as cognitive psychology,

neuroscience, personality, social psychology, and (cross cul-

tural psychology. Today we have a wealth of knowledge about

basic psychological phenomena, and the mission of user psy-

chology is to organize and translate the various types of know-

ledge into coherent wholes that can work as a reference in

solving problems. Alas, we do not yet have a unified psychol-

ogy, and psychological research areas are very different from

each other. For example, the limited capacity of working mem-

ory can be used to solve some interaction questions, while the

psychology of emotions is applicable to very different types of

problems. It is true that these fields have some connection points

such as the Yerkes-Dodson’s law, but still the two frameworks

can and should be used in solving very different types of prob-

lems. Consequently, it would be self-handicapping to adopt as

the main goal a unified theory of user psychology. User psy-

chological explanatory frameworks must be and are being built

on the basis of many types of psychological knowle dge.

Conclusion: The New Role of Psychology in I CT

The sheer scale of technological development over the mil-

lennia has been astonishing. Once-innovative technologies be-

came either obsolete or mundane elements of everyday con-

temporary life as human societies progressed. The technologi-

cal developments of especially the last decade have dramati-

cally altered most humans’ way of life. Interaction with objects

and people is not direct anymore, if it ever was, but increasing-

ly requires interaction with computers manifested as hardware

of all kinds and, more recently, software. It would not be an

exaggeration to view humans as standing at the cusp of pro-

found social changes that are in line with those following the

invention of writing or the steam engine. Therefore, now is a

good time to stop for a moment and ponder the forces that make

such new developments possible.

For decades it has been sufficient to intuitively imagine hu-

man interaction with new services. Today, however, it is ne-

cessary to admit that mere intuitions have not given us a suffi-

ciently deep enough understanding of the human mind and

social roles to enable us to design really good technical prod-

ucts for people. Scientists working with the human mind and

society have naturally known for a long time that simple intui-

tions and lay science do not provide technologists with an ac-

curate understanding of people. This is why psychology, soci-

ology, anthropology, and other human sciences have developed

sophisticated observations, concepts, and theories that make it

easier for specialists to understand better human being.

We have proposed user psychology as an explanatory prob-

lem-solving science contributing to the development and so-

cietal use of interactive technology. A hundred and fifty years

of scientific psychology has developed a deep understanding of

the regularities and principles of human mind. We understand

much about emotions, motives, thinking, psychology of lan-

guage and social processes. To an outsider, it may seem un-

P. SAARILUOMA & A. OUL ASVIR TA

326

problematic to expect that psychological knowledge in all of its

breadth and depth should form a basis for modern information

technology and that interaction should be designed on the ground

of professional psychology. Nevertheless, the difficulties in

applying psychology in industry have been known for a long

time. The links between psychological knowledge and con-

structive processes have not met each other in an ideal manner

(Carroll, 1991, 1997). The purpose of this paper has been to

search for solutions to some foundational issues that we think

underlie these problems. We have repurposed the classic con-

cept of user psychology, suggesting a number of new elabora-

tions to it (Moran, 1981).

The need to reformulate user psychology has arisen, because

original formulations have been found limited and no compre-

hensive alternatives have been proposed. The information

processing model of human cognition is mechanical, because it

assumes that each act is a consequence of the prevailing states

and state histories. The activity theory and related ideas have

not been able to host contemporary psychological work due to

their unwillingness to accept internal mechanisms of mind as

the explanatory base. Building on earlier mainstream ontologies,

we have taken the human being as an intentional creature to the

subject of inquiry in this discipline. Instead of activity, we talk

about intentional action with interactive technology.

By doing this, we have tried to address two unfortunate con-

sequences of the previous formulations of psychological re-

search in this area. First, it has been recognized that psycholog-

ical research has fragmented into niches that—notwithstanding

few promising exceptions—have been defined by technological

boundaries (Carroll, 1997; Olson & Olson, 2003). User psy-

chology should be formulated around psychological questions

so that we can merge these fragments over artificial boundaries.

This is the essence of what “putting the user first” means for a

scientist. Secondly, user psychology is needed for replacing the

folk-psychological thinking by scientific psychology and its

principles. Current folk-psychological practice simply does not

accumulate—the same things are being repeated in a frustrating

cycle. Explanatory practices should replace seat-of-the-pants

thinking. However, finding explanations to particular problems

can also be regarded as search for the most suitable form of

psychological knowledge.

Thus far, the development of technology has belonged to en-

gineers and computer scientists, who relatively seldom have

in-depth knowledge of psychological phenomena and know-

ledge. Consequently, psychological knowledge has been

brought to the picture only after devices or platforms have al-

ready been built. The question is if we could do better. We

expect user psychology to contribute to earlier stages of devel-

opment. For example, usability research with its interest in

immediate suitability of technology for human action is ulti-

mately a conservative form of possible psychological perspec-

tives. What we could do is to turn this thinking upside-down by

first designing human actions and after that asking what kinds

of technologies would be useful. At first blush, this does not

sound like news to researchers trained in user-centered de-

s ign—it is known that the earlier psychological knowledge can

be taken as a part of action analysis, the better human require-

ments can be implemented to interaction. However, we claim

that user psychology can a) do this work more systematically

than approaches that rely on folk psychology and b) it can help

us systematically examine the very premise that technology is

needed in the first place.

However, we do not expect user psychology to be easy. In

the short term, it is time-consuming to deal with psychological

knowledge compared to running a simple study. But this is the

only way out of the method of trial and error. Moreover, getting

psychologists interested in “applied psychology” may be diffi-

cult. Relatively few psychologists are seriously interested in

“technical issues” as they have hands full with traditional prob-

lems within departmental focus areas, such as education, mental

illnesses, family violence etc. However, we see the possible

rewards motivating enough that solutions can be found. Tech-

nology has become so pervasive an aspect of modern life that

its relationship with human mind deserves the status of a basic

research question and its own discipline. Psychology of com-

puter use should not be the handmaiden of technologists but

define itself by its own terms.

Re ferences

Aaker, J. (1997). Dimensions of brand personality. Journal of Market-

ing Research, 34, 347-356.

Anderson, J. (2002). ACT: A simple theory of complex cognition. Cam-

bridge: MIT Press.

Annett, J. & Duncan, K. (1967). Task analysis and training design.

Journal of Occupational Psychology, 41, 211-221.

Anderson, J., Farrell, R., & Sauers, R. (1984). Learning to program in

LISP. Cognitive Science, 8, 87-129.

Backstrom, L., Huttenlocher, D., Kleinberg, J., & Lan, X. (2006).

Group formation in large social networks: membership, growth, and

evolution. In Proceedings of the 12th ACM SIGKDD international

conference on knowledge discovery and data mining (pp. 44-54).

New York: ACM Press.

Baddeley, A. (1986). Working memory. Oxford, UK: Oxford University

Press.

Bannon, L. (1991). From human factors to human actors: The role of

psychology and human-computer interaction studies in system design.

In J.M. Greenbaum & M. Kyng (Eds.), Design at work: cooperative

design of computer systems (pp. 25-44). Lawrence Erlbaum Asso-

ciates.

Bannon, L. & Bødker, S. (1991). Beyond the interface: Encountering

artifacts in use. In Carroll, J. (Ed.), Designing interaction: psycholo-

gy at the human-computer interface. Cambridge: Cambridge Univer-

sity Press.

Beck, A. (1976). Cognitive therapy and the emotional disorders. Wis-

consin: International Universities Press.

Beenen, G., Ling, K., Wang, X., Chang, K., Frankowski, D., Resnick,

P., & Kraut, R. (2004). Using social psychology to motivate contri-

butions to online communities. In Proceedings of the 2004 ACM

Conference on Computer Supported Cooperative Work (pp. 212-221).

New York, NY: ACM Press.

Bernstein, R. (1971). Praxis and action:contemporary philosophies of

human activity. Pennsylvania, PA: University of Pennsylvania Press.

Blomberg, J., Burrell, M., & Guest, G. (2002). An ethnographic ap-

proach to design. In The human computer interaction handbook:

fundamental, evolving technologies and emerging applications (pp.

964-986). Hillsdale, NJ: Lawrence Erlbaum Associates Inc.

Bødker, S. (1989). A human activity approach to user interfaces. Hu-

man-Computer Interaction, 4, 171-195.

Bourges-Waldegg, P. & Scrivener, S. (1998). Meaning, the central

issue in cross-cultural HCI design. Interacting with Computers, 9,

287-309.

Broadbent, D. (1958). Perception and communication. London, UK:

Pergamon Press.

Card, S. (2003). Information visualization. In The human-computer

interaction handbook: fundamentals, evolving technologies and

emerging applications (pp. 544-582). Hillsdale, NJ: Lawrence Erl-

P. SAARILUOMA & A. OUL ASVIR TA

327

baum Associates Inc.

Card, S., Moran, T., & Newell, A. (1983). The psychology of human-

computer interaction. Hillsdale, NJ: Lawrence Erlbaum Associates

Inc.

Carroll, J. (1991). Introduction: The Kittle House Manifesto. In J. Car-

roll (Ed.), Designing interaction: psychology at the human-computer

inte r face. New York, NY: Cambridge University Press.

Carroll, J. (1997). Human-computer interaction: Psychology as a

science of design. Annual Review of Psychology, 48, 61-83.

Carroll, J. (2000). Making use: scenario-based design of humancomp u-

ter interactions. Cambridge, MA: MIT Press.

Chase, W. & Simon, H. (1973). Perception in chess. Cognitive Psy-

chology, 4, 55-81.

CHI Conference (2009). Guide to a Successful HCI Archive Submis-

sion. URL (last checked 15 May 2009) http:// w w w .chi 2009.org/Au-

thors/Guides/Ar c hiveGuide.html

Choi, B., Lee, I., Kim, J., & Jeon, Y. (2005). A qualitative cross-na-

tional study of cultural influences on mobile data service design. In

Proceedings of the SIGCHI conference on Human factors in compu-

ting systems (pp. 661-670). New York, NY: ACM Press.

Clemmensen, T. (2006). Whatever happened to the psychology of hu-

man-computer interaction? Information Technology & People, 19,

121-151.

Cockton, G. (2008). Designing worth-connecting preferred means to

desired ends. Interactions, 15, 54-57.

Cole, M. (1991). On cultural psychology. American anthropologist, 93,

435-439.

Coren, S., Ward, L., & Enns, J. (1999). Sensation and perc ep tion 5th ed.

Fort Worth, TX: Harcourt Brace.

Cowan, N. (2001). The magical number 4 in short-term memory: A

reconsideration of mental storage capacity. Behavioral and Brain

Sciences, 24, 87-114.

Czaja, S., Charness, N., Fisk, A., Hertzog, C., Nair, S., Rogers, W., &

Sharit, J. (2006). Factors predicting the use of technology: Findings

from the center for research and education on aging and technology

enhancement (CREATE). Psychology and Aging, 21, 333.

Dennett, D. (1991). Consciousness explained. London, UK: The Pen-

guin Press.

Diaper, D. & Stanton, N. (2004). The handbook of task analysis for

human-computer interaction. Mahwah, NJ: Lawrence Erlbaum As-

sociates Inc.

Dourish, P. (2001). Wh ere the action is: the foundations of embodied

inte r action. Cambridge, MA: MIT Press.

Dreyfus, H. (1992). What computers still can’t do: a critique of artifi-

cial rreason. Cambridge, MA: MIT Press.

Ekman, P. (2003). Emotions revealed. New York, NY: Times Books.

Ericsson, K. & Simon, H. (1993). Protocol an alysis. Cambridge, MA:

MIT Press.

Fitts, P. (1954). The information capacity of the human motor system in

controlling the amplitude of movement. Journal of Experimental

Psychology, 47, 381-391.

Forlines, C., Shen, C., Wigdor, D., & Balakrishnan, R. (2006). Explor-

ing the effects of group size and display configuration on visual

search. In Proceedings of the 2006 20th anniversary conference on

Computer supported cooperative work (pp. 11-20). New York, NY:

ACM Press.

Friedlander, N., Schlueter, K., & Mantei, M. (1998). Bullseye! when

Fitts’ law doesn’t fit. In Proceedings of the SIGCHI conference on

Human factors in computing systems (pp. 256-264). New York, NY:

ACM Press.

Frieze, C., Hazzan, O., Blum, L., & Dias, M. (2006). Culture and envi-

ronment as determinants of women's participation in computing: re-

vealing the women-CS fit. ACM SIGCSE Bulletin, 38, 22-26.

Gadamer, H. (1960). Wahrheit und methode. Mohr.

Goldstone, R. (1998). Perceptual learning. Annual Review of Psycholo-

gy, 49, 585-612.

Gray, W. (Ed.) (2007). Integrated models of cognitive systems (cogni-

tive models and acrhitectures). Oxford, UK: Oxford University Pre-

ss.

de Groot, A. (1965). Thought and choice in chess. The Hague, Nether-

lands: Mouton.

de Groot, A. (1966). Perception and memory versus thought: Some old

ideas and recent findings. In B. Kleinmuntz (Ed.), Problem solving:

research, methods, and theory. Wiley.

Grossmann, R. (1984). Phenomenology and existentialism: an intro-

duction. Routledge & Kegan Paul Books.

Grudin, J. (1990). The computer reaches out: the historical continuity of

interface design. In Proceedings of the SIGCHI conference on Hu-

man factors in computing systems: Empowering people (pp.

261-268). New York, NY: ACM Press.

Grudin, J. (1994). Computer-supported cooperative work: History and

focus. Computer, 27, 19-26.

Hawthorn, D. (2000). Possible implications of aging for interface de-

signers. Interacting with Computers, 12, 507-528.

Heath, C. & Luff, P. (2000). Technology in action. Cambridge, UK:

Cambridge University Press.

Heidegger, M. (1927/1962). Being and time. London, UK: SCM Press.

Helander, M. (1997). Forty years of idea: some reflections on the evo-

lution of ergonomics. Ergonomics, 40 (10), 952-961.

Helander, M., Landauer, T., & Prabhu, P. (1997). Handbook of human-

computer interaction. New York, NY: Elsevier Science Inc.

Hofstede, G. (1984). Culture ’s consequences: International differences

in work-related values. London, UK: Sage.

Hogan, R., Harkness, A., & Lubinski, D. (2000). Personality and indi-

vidual differences. In K. Pawlik & M.R.Rosenzweig (Eds.) Inte r na-

tional handbook of psychology (pp. 283-304). London, UK: Sage.

Horvitz, E. (1999). Principles of mixed-initiative user interfaces. In

Proceedings of the SIGCHI conference on Human factors in compu-

ting systems: the CHI is the limit (pp. 159-166). New York, NY:

ACM Press.

Husserl, E. (1913/1982). Ideas pertaining to a pure phenomenology

and to a phenomenological philosophy—first book: general intro-

duction to pure phenomenology. Trans. F. Kersten. The Hague,

Netherlands: Nijhoff.

Ishii, H. & Ullmer, B. (1997). Tangible bits: towards seamless interfac-

es between people, bits and atoms. In Proceedings of the SIGCHI

conference on Human factors in computing systems (pp. 234-241).

New York: ACM Press.

Isler, R., Kirk, P., Bradford, S., & Parker, R. (1997). Testing the rela-

tive conspicuity of safety garments for new zealand forestry workers.

Applied Ergonomics, 28, 323-329.

John, B.E., Prevas, K., Salvucci, D.D., & Koedinger, K. (2004). Predic-

tive human performance modeling made easy. In Proceedings of the

SIGCHI conference on human factors in computing systems(pp.

455-462). New York, NY: ACM Press.

Jordan, P. (1998). Human factors for pleasure in product use. Applied

Ergonomics, 29, 25-33.

Kaptelinin, V. (1996). Activity Theory: Implications for Human Com-

puter Interaction In B. Nardi (Ed.), Context and Consciousness: Ac-

tivity Theory and Human-Computer Interaction (pp. 103-116). Cam-

bridge, MA: MIT Press.

Kirwan, B. & Ainsworth, L. (1992). A guide to task analysis. UK:

Taylor & Francis.

Kuutti, K. (1996). Activity theory as a potential framework for hu-

man-computer interaction research. In B. Nardi (Ed.), Context and

consciousness: activity theory and human-computer interaction (pp.

17-44). Cambridge, MA: MIT Press.

Kuutti, K. & Kaptelinin, V. (1997). Rethinking cognitive tools: From

augmentation to mediation. In Proceedings of the Second Interna-

tional Conference on Cognitive Technology: Humanizing the Infor-

mation Age , Japan, pp.31-32.

Laird, J., Newell, A., & Rosenbloom, P. (1987). SOAR: An architecture

for general intelligence. Artificial intelligence, 33, 1-64.

Landauer, T.K. (1987). Relations between cognitive psychology and

computer system design. In J. Carroll (Ed.), Interfacing thought:

cognitive aspects of human-computer interaction (pp. 1-25). Cam-

bridge, MA: MIT Press.

Laudan, L. (1977). Progress and its problems. Berkeley, CA: Univer-

P. SAARILUOMA & A. OUL ASVIR TA

328

sity of California Press.

Leontyev, A. (1981). Problems of the development of the mind. Mos-

cow, Russia: Progress.

Lindzey, G. & Borgatta, E. (1954). Sociometric measurement. In Hand-

book of social psychology (pp. 405-448). Addison-Wesley Publishers

Co.

Mahner, M. (2001). Scientific realism: Selected essays of mario bunge.

Amherst, NY: Prometheus Books.

Malone, T.W. & Crowston, K. (1994). The interdisciplinary study of

coordination. ACM Computing Surveys, 26, 87-119.

Martin, J. (1973). Design of man-machine dialogues. Pre ntice-Hall.

Matsumoto, D. (1996). Culture and psychology. Pacific Grove, CA:

Brooks/Cole.

McCarthy, J. & Wright, P. (2004). Technology as experience. interac-

tion s , 11, 42-43.

McCrae, R. & Costa Jr., P. (1997). Personality trait structure as a hu-

man universal. American Psychologist, 52, 509-516.

McKeithen, K., Reitman, J., Rueter, H., & Hirtle, S. (1981). Knowledge

organization and skill differences in computer programmers. Cogni-

tive Psychology, 13, 307-325.

Merleau-Ponty, M. (1962). Phenomenology of Perception. Routledge &

Kegan Paul.

Meyer, D. & Kieras, D. (1997). A computational theory of executive

cognitive processes and multiple-task performance: Part 1. basic

mechanisms. Psychological Review, 104, 3-65.

Miller, G. (1951). Language and Communication. New York, NY:

McGraw-Hill.

Moran, T. P. (1981). Guest editor's introduction: An applied psycholo-

gy of the user. ACM Computing Surveys, 13, 1-11.

Nardi, B. (1996) (Ed.). Context and Consciousness: Activity Theory and

Human-Computer Interactio n. Cambridge, MA: MIT Press.

Newell, A. & Card, S. (1986). Straightening out softening up: Response

to Carroll and Campbell. Human-Computer Interaction, 2, 251- 267.

Newell, A. & Simon, H. (1972). Human problem solving. Englewood

Cliffs, NJ: Prentice-Hall.

Nielsen, J. (1993). Usability Engineering. Morgan Kaufmann.

Norman, D. (2004). Emotional design: why we love (or hate) everyday

things. New York: Basic Books.

Norman, D. & Draper, S. (1986) (Eds.). User centered system design;

new perspectives on human-computer interaction. Hillsdale, NJ:

Lawrence Erlbaum Associates Inc.

Norman, D. A. (1987). Some observations on mental models. In W.A.S.

Buxton & R.M. Baecker (Eds.), Human-computer interaction: a

multidisciplinary approach (pp. 241-244). San Francisco, CA: Mor-

gan Kaufmann Publishers Inc.

Olson, G. & Olson, J. (2003). Human-computer interaction: Psycho-

logical aspects of the human use of computing. Annual Review of

Psychology, 54, 491-516.

Pahl, G. & Beitz, W. (1996). Engineering design: a systematic ap-

proach. London: Springer.

Parasuraman, R. & Rizzo, M. (2006). Neuroergonomics: The Brain at

Wo rk. Oxford, UK: Oxford University Press.

Pashler, H. (1998). The Psychology of attentio n. Cambridge, MA: MIT

Press.

Picard, R.W. (1997). Affective computing. Cambridge, MA: MIT Press.

Power, M. & Dalgleish, T. (2007). Cognition and emotion: from order

to disorder. New York: Psychology Press.

Rauterberg M. (2006). From personal to cultural computing: how to

assess a cultural experience. In G. Kempter & P. von Hellberg (Eds.),

uDayIV--information nutzbar machen (pp. 13-21). Lengerich: Pabst

Science Publisher.

Rosson, M.B. & Carroll, J. (2002). Usability engineering: scenario-

based development of human-computer interaction. San Francisco,

CA: Morgan Kaufmann Publishers Inc.

Roth, W. & Lee, Y. (2007). Vygotsky’s neglegted legacy: Cultura l-

historical activity theory. Review of Educational Research, 77, 186.

Saariluoma, P. (1997). Foundational analysis: presuppositions in expe-

rimental psychology. New York: Routledge.

Saariluoma, P. (2005). Explanatory frameworks for interaction design.

In. A. Pirhonen, P. Saariluoma, H Isomaki, & C. Roast (Eds.) Future

interaction design (pp. 67-83). London: Springer.

Sackman, H. (1970). Man-Computer Problem Solving. Princeton, NJ:

Auerbach.

Schraagen, J. & Chipman, S. (2000). Cognitive Task Analysis. NJ:

Lawrence Erlbaum Associates Inc.

Searle, J. (1983). Intentionality, an essay in the philosophy of mind.

Cambridge, UK: Cambridge University Press.

Sellen, A., Rogers, Y., Harper, R., & Rodden, T. (2009). Reflecting

human values in the digital age. Communications of the ACM, 52,

58-66.

Shackel, B. (1959). Ergonomics for a computer. Design, 120, 36-39.

Shneiderman, B. (1976). Exploratory experiments in programmer be-

havior. International Journal of Parallel Programming, 5, 123- 143.

Simon, H. (1969). The sciences of the artificial. MIT Press, Cambridge,

MA.

Styles, E. (1997). The psychology of attention. New York: Psychology

Press.

Suchman, L. (1987). Plans and situated actions: the problem of hu-

man-machine communication. Cambridge, UK: Cambridge Univer-

sity Press.

Svanaes, D. (2001). Context-aware technology: a phenomenological

perspective. Human-Computer Interaction, 16, 379-400.

The Joint Accident Investigation Commission of Estonia, F. and Swe-

den (1997). Final report on the capsizing on 28 September 1994 in

the baltic sea of the RO-RO passenger vessel MV Estonia. Technical

report, The Joint Accident Investigation Commission.

Ulrich, K. & Eppinger, S. (2003). Product design and development.

New York, NY: McGraw-Hill.

Vertegaal, R. (2003). Attentive user interfaces: Introduction. Commu-

nications of the ACM, 46, 30-33.

Von Wright, G. (1972). On so-called practical inference. Acta Sociolo-

gic a, 15, 39-53.

Vygotsky, L. (1980). Mind in society. Cambridge, MA: Harvard Uni-

versity Press.

Weinberg, G. (1971). The psychology of computer programming . New

York, NY: Van Nostrand Reinhold.

Weiser, M. (1991). The computer for the twenty-first century. Scientific

Amer ic an, 265,94-104.

Welford, A. (1968). Fundamentals of skill. Methuen.

Wickens, C., Lee, J., Liu, Y., & Gordon-Becker, S. (2003). Introduc-

tion to human factors engineering. Upper Saddle River, NJ: Prentice -

Hall, Inc.

Wiggins, J. (1996). The five-factor model of personality: Theoretical

perspectives. New York, NY: Guilford Press.

Winograd, T. & Flores, F. (1986). Understanding computers and cog-

nitio n. Norwood, NJ: Ablex Publishing Corp.

Wittgenstein, L. (1958). The blue and brown books. Oxford, UK:

Blackwell.