The Beneﬁts of Considering Psychological Reactance as Users’

Personality Trait in HCI Research

Proﬁling Users with Hong’s Psychological Reactance Scale

Patrick Ehrenbrink

Quality and Usability Lab, Technische Universitt Berlin, Ernst-Reuter-Platz 7, Berlin, Germany

Keywords:

Psychological Reactance, User Models, Personality Traits, Adaptation.

Abstract:

This paper argues for considering psychological reactance as a personality trait as an aspect for user models

in the ﬁeld of human-computer interaction (HCI). It is pointed out that with increasing human-likeness of

services, such as intelligent personal assistants, phenomena from human-human interaction are becoming

more important in the ﬁeld of HCI, as well. The concept and means of measuring psychological reactance are

explained. Furthermore, methods for adapting to different levels of trait reactance are proposed and discussed.

1 INTRODUCTION

Usability and user experience research has success-

fully established a set of metrics and requirements for

the assessment and design of human-computer inte-

raction for years. For example EN ISO 9241-11 sta-

tes effectiveness, efﬁciency and user-satisfaction as

the main criteria for usability (ISO 9241-11, 1999).

This criteria are perfectly valid and have been so for

decades and for most machines, devices and servi-

ces. Modern services, such as Apples Siri (Apple

Inc., 2011) however, have introduced a new aspect to

human-machine interaction: human-likeness.

Even though there have been anthropomorphous ro-

bots and services that were intended to act human-

like before, they have usually lacked the level of

artiﬁcial intelligence and ﬂexibility that would have

been required to be really convincing. Modern intelli-

gent personal assistants (IPAs) have reached a level of

complexity and naturalness of interaction that has not

been possible before. Several features and aspects of

interaction that have been implemented into the most

popular IPAs (Siri, Cortana (Microsoft, 2014), Goo-

gle Now (Google Inc., 2012)) are probably responsi-

ble for this. Six of those features are listed below, but

do not necessarily represent the complete list.

Natural Language Understanding. The ﬁrst one,

natural language understanding, can greatly improve

the naturalness of human-computer interaction be-

cause it allows users to interact with the device in the

same modality, as they do with other humans in their

everyday life

. While natural language interaction is

not always the most precise or fastest interaction mo-

dality, it has become more important with the increa-

sed use of small mobile devices, that have the inherent

lack of effective input devices, such as (large) keybo-

ard and mouse. Also, natural language understanding

provides hands-free interaction for situations in which

the user might not be able or willing to use other me-

ans of interaction, such as when driving a car.

Personality. Siri and Cortana have a their own per-

sonality that is reﬂected in their behavior and their

voice. Both have human-like voices that can easily be

assigned to a speciﬁc gender. Cortana is often asso-

ciated with a face that initially appeared in the com-

puter game ”Halo” (faturing a holographic, ﬁctional

version of cortana). On requests, both IPAs are able

to tell nerdy jokes or talk about their private intere-

sts. Additionally, they can get bored or annoyed by a

conversation and respond with irony or switching the

topic.

Adaptivity. Adaptivity is a feature that enables an

IPA (or any other service) to adapt to its environment

or its user. For example, Google Now can analyze

its user’s browsing behavior, identify topics of inte-

rest and then recommend new web-content that ﬁts

Even though other modalities, such as gestures, gaze

etc. take a big part in this, as well, spoken language is the

most salient one.

158

Ehrenbrink P.

The Beneﬁts of Considering Psychological Reactance as Usersâ

Z Personality Trait in HCI Research - Proﬁling Users with Hongâ

Zs Psychological Reactance Scale.

DOI: 10.5220/0006265301580164

In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2017), pages 158-164

ISBN: 978-989-758-229-5

the assumed interests of the user. Also, IPAs usually

take environmental information, such as the location,

into account for recommendations of e.g. restaurants.

This gives the IPAs a rough awareness of the con-

text of the interaction, at least in the impression of

the user.

Proactive Behavior. Proactive behavior is often

combined with adaptation to the user or the environ-

ment. The IPA can use data of its user in order to

actively recommend content or services to the user.

Some example of Google’s Now are: The service can

infer from moving behavior and other information,

where its user works and offers navigation for the fas-

test way home on its dashboard when work is ﬁnished

(or it inferred, that work should have ﬁnished). Also,

it can ﬁnd boarding cards in the user’s emails and

present it proactively (in the dashboard) on boarding

time. Proactive behavior is usually the expression of

an agenda or intend and therefore probably a feature

that adds strongly to the perceived human-likeness of

IPAs. But proactive behavior implies that the initi-

ative for an action comes from the system, thereby

taking away some control over the situation from the

user. This makes proactive behavior especially criti-

cal for highly reactant users, as explained later on.

Complex Feature Set. Sophisticated IPAs are not

only recommendation and conversation services, they

also feature a wide set of other functions. Thereby,

they act as an interface to other apps of the smart

phone or even devices of the smart home with which

they are connected. E.g. the feature set of Siri inclu-

des managing calendar entries, writing emails, wri-

ting notes, open calls, control smart home devices

and answer all kinds of requests from fact queries to

singing songs. The complex set of feature makes it

unpractical to thoroughly evaluate such a service in

terms of its usability properties, because all functions

would have to be evaluated under reliable conditions.

Also, it becomes impossible or at least not reasona-

ble for the user to grasp the complete functionality of

such a service. This is even ampliﬁed by the fact that

IPAs usually do not work locally but are web-based

services and new functionalities can be implemented,

changed or removed without knowledge of the user.

The uncertainty about the abilities of a service adds

to the human-likeness of interaction because the user

is not able to fully understand the limitations of that

service. This might imply an intelligent interaction

partner to the human.

Hierarchy. Humans are used to be in a supreme po-

sition towards the devices that they use. In the tra-

ditional interaction paradigm, humans are taking the

initiative and control the machines that they interact

with. In modern IPAs and some other services and

devices, this paradigm is not fully applicable, any-

more. Such services or devices sometimes take the

initiative themselves, or autonomously adapt to chan-

ges in the situation. They express a certain level of

autonomy, which might reduce the hierarchy gap bet-

ween the service or device and the user.

The above stated features of IPAs and some other

services and devices are reasons that make a broad,

systematic usability evaluation with traditional met-

hods a hard task. The results will be difﬁcult to in-

terpret and reliability will be poor, because users will

react very diverse to many of those features. Just as

in human-human interaction, factors that are not mere

performance indices, such as behavioral patterns of

the service or social presence of an IPA might play an

increasing role alongside increasing human-likeness.

For example, a study with elderly people observed

that users’ reactions towards persuasive attempts are

more positive if the persuasive agent showed social

behavior, such as humans do (Looije et al., 2010).

On the other hand it was observed that highly cons-

cientiousness people like agents that show social be-

havior less, compared to other people (Looije et al.,

2010). In order to gain a more complete and meaning-

ful overview of such a system for both, sumative and

formative evaluation, new metrics are needed that can

access affective user states and personality traits and

thereby help to build heuristics about a user’s prefe-

rences. In this paper, it is argued, that such heuristics

can lead to more acceptance among users.

There are multiple character traits and affective

states that potentially play a role in the interaction be-

tween humans and human-like devices and services.

Among those, probably the big ﬁve personality traits

(Goldberg, 1993), that were used by Looije et al. to

investigate users’ preferences for agent based intelli-

gent assistants (Looije et al., 2010), are the most com-

monly known. For this work, however, reactance as a

personality trait is in focus. This is due to reactance

going along with diminished acceptance (Dillard and

Shen, 2005; Ehrenbrink et al., 2016a), what proba-

bly makes reactance one of the more important and

interesting factors for HCI developers. Further, the

discussion will mostly include reactance as a persona-

lity trait, in contrast to reactance as an affective state.

Personality traits can be used to create a personality

proﬁle of a user which would need to be created only

once and would rarely require actualization. The as-

sessment of users’ state reactance would require con-

stant or very frequent measurements and can be con-

sidered as impractical for the lack of appropriate me-

The Beneﬁts of Considering Psychological Reactance as Usersâ

Z Personality Trait in HCI Research - Proﬁling Users with Hongâ

Psychological Reactance Scale

159

asurement techniques. During further discussion, se-

veral examples on how to employ reactance measure-

ment in HCI will be given. As the aim of this paper

is to explain the concept of reactance and to argue for

its beneﬁts and possible applications in HCI research

and design, only two simple use-cases will be given.

The seat belt alarm of cars and a ﬁtness application

utilize basic interfaces and their functionality is not

complex. Here, only few factors apart from reactance

have to be considered. This makes those use-cases

easier to explain, compared to a full-scale IPA.

2 PSYCHOLOGICAL

REACTANCE

Psychological reactance is a concept from psycho-

logy that was initially introduced by (Brehm, 1966).

Brehm described reactance as a motivational state that

effects a persons behavior towards a perceived threat

to the personal freedom or freedom of choice. The-

refore, a person that has received, or thinks to have

received, a freedom threat will become reactant. As a

consequence, the option that has been eliminated by

the freedom thread will appear more attractive to that

person as it would have been otherwise (Brehm and

Brehm, 1981). On the other hand, the source of the

freedom threat, e.g. the person who denied an certain

freedom will be downgraded in its judgement and ap-

pear more negative (Brehm and Brehm, 1981).

Also, that person will experience a cognitive dis-

sociation because he or she wants to exercise a cer-

tain freedom but is forbidden or unable to do so. In

order to then solve the cognitive dissociation, the per-

son will try to restore the threatened freedom. This

can be done either directly, or indirectly (Brehm and

Brehm, 1981). A direct attempt to restore the threa-

tened freedom would be to directly exercise the thre-

atened freedom. For example, if a child is denied a

cookie, it might wait until the parents leave the room

and then get a cookie even though it is forbidden. Di-

rectly restoring a threatened freedom is however often

regarded as antisocial and therefore, the cognitive dis-

sonance is often resolved indirectly. This can be done

by exercising another freedom that is not threatened

but somehow similiar to the threatened one. Even

though that other freedom was not threatened, exer-

cising it can restore the freedom that was lost (Brehm

and Brehm, 1981). The child whose parents denied

it a cookie would then, instead of secretly eating a

cookie, secretly eat some grapes and by that recreate

its freedom.

Finally, a coping strategy that can be exercised is

denial of the freedom threat. This can often be the

case if the freedom threat is ambiguous or unclear

(Brehm and Brehm, 1981). Denial of the freedom

threat has the advantage that a freedom restoration is

not necessary. Therefore, simply ignoring or denying

the threat, without further acting against the prohibi-

tion (because this could invoke unwanted consequen-

ces) can be an easy and effective way to handle the

cognitive dissonance.

Reactance can be a motivational or affective state

that is time-variant and dependent on the situation.

But reactance can also be a personality trait. Trait

reactance can be regarded as a person’s proneness to

become reactant. A person, who’s personality is very

reactant will also be more likely to enter a reactant

state and then exercise reactant behavior as it is des-

cribed above. If for example a service or device indu-

ces reactance in its user, it is likely to suffer negative

consequences to its acceptance or its persuasive inﬂu-

ence over the user. At this point, it is important to note

that personality traits, such as trait reactance are rela-

tively stable over time and therefore would not have

to be assessed regularly.

3 MEASURING TRAIT

REACTANCE

Trait reactance can be measured with questionnaires.

Several questionnaires have been developed that me-

asure trait reactance, mostly by a series of self report

items. The ﬁst one was the ”Fragebogen zur Messung

der psychologischen Reaktanz”, published by (Merz,

1983). Merz’s questionnaire was later translated and

adapted by (Hong and Page, 1989; Hong, 1992; Hong

and Faedda, 1996). ”Hong’s Psychological Reactance

Scale” has been validated (Hong, 1992) and discussed

in therms of its psychometric properties (Jonason and

Knowles, 2006; Shen and Dillard, 2010). Findings

in literature support the possibility to treat Hong’s

Psychological Reactance Scale as an unidimensional

scale. It has eleven self-report items that are answered

on a Likert-type scale (Likert, 1932). The relatively

small size of the questionnaire and the simple analy-

sis of its results make it an easy-to-use tool to quickly

assess a person’s trait reactance.

Assessing users’ trait reactance requires them to

ﬁll out Hong’s Psychological Reactance Scale (or

another appropriate one). While this does not take

long, there is still the hurdle of handing out sensitive

information. The act of asking users to give such in-

formation might trigger strong reactance itself, which

could result in opting out to give such information or

in giving false information. To avoid this, it should

be clearly stated, why such information is collected

HUCAPP 2017 - International Conference on Human Computer Interaction Theory and Applications

160

and how the user will beneﬁt from that. Research

has shown that reactance to the use of personal data

decreases with justiﬁcation for the collection of such

data and with utility (White et al., 2008). As persona-

lity traits usually do not change much, trait reactance

would have to be assessed either rarely or even only

once and then be stored in the user proﬁle.

Knowledge about a person’s level of trait reac-

tance can be usefull because interaction can then be

tailored and adapted to ﬁt the persons personality pro-

ﬁle. Also, evaluation results can be interpreted with

the persons personality traits, here, trait reactance in

mind. For example, a recent (not yet published) eva-

luation of Apple’s Siri, using the AttrakDiff Mini

(Hassenzahl and Monk, 2010) questionnaire, revealed

that highly reactant persons rate Siri’s hedonistic and

pragmatic quality worse than people who show low

levels of trait reactance. The results reached signiﬁ-

cant levels t(22)=2.27, p=0.03, d=0.927 for hedonistic

quality, but failed to do so for pragmatic quality, with

d = 0.496 (N=24).

4 UTILIZING REACTANCE

RESEARCH FOR HCI

4.1 Avoiding Reactant Behavior

Multiple techniques exist that are able to decrease re-

actance effects. While it can be adequate to always

use those techniques for many situations, for exam-

ple for non-adaptive systems or if no user information

is known, a user-speciﬁc tuning of the trade-off bet-

ween the users’ reactance and frequency and magni-

tude of system-persuasion could often be used to opti-

mize compliance. To further clarify this, an example

on how the seat belt alarm in a car could be impro-

ved is given for each technique. The seat belt alarm

is a particularly ﬁtting use case, because it is a persu-

asive mechanism that car manufacturers use to make

drivers wear seat belts. Because of the importance of

wearing seat belts, the alarm is usually designed to be

quite annoying for the driver. The intention behind

this is to make the driver fasten the seat belt as soon

as possible. The downside of this is that many drivers

get reactant and try to bypass the seat belt alarm while

refraining from fastening the seat belt.

Magnitude of Threat. The level of reactance that

is induced is also dependent on the magnitude of the

threat or request. This means that if a freedom threat

or a request is affecting a person a lot, this will cause

more reactance compared to a freedom threat or re-

quest that has little effect on a person (Brehm and

Brehm, 1981). With respect to reactance as a persona-

lity trait, one can expect that a highly reactant person

is also more sensitive to such requests. A seat belt

alarm for a highly reactant person could likely be im-

proved if it would involve a message like ”Please put

the seat belt on, its not a hard thing to do.”. In the

mobile context, it could be wise for a service or soft-

ware not to request additional access rights or present

lengthy forms to ﬁll out to users that are known to be

highly reactant, because the result could be either the

discontinuation of usage or diminished acceptance. In

the case of forms, a service might also receive in-

tentionally false information because a highly reac-

tant person attempted to restore personal freedom by

acting against the freedom threat.

Magnitude of Request. The magnitude of request

is moderated by the explicitness of the request. This

means that a formulation like ”Fasten your seat belt!”

is a strong magnitude of request, where emphasizes

lies on the request. Whereas ”Fastening your seat belt

can signiﬁcantly improve your chances of survival in

an accident.” might transport the same message, na-

mely that someone should fasten the seat belt, but em-

ploys a more indirect persuasion strategy. The use of

an objective argument against a command can incre-

ase persuasive power by evoking intrinsic motivation

and avoiding reactance. There has been a number of

studies that show that magnitude of request can mode-

rate compliance to the message (Rains, 2013; Silvia,

2005).

But You Are Free to Accept or Refuse Technique.

A rather simple method to reduce reactant behavior

and to increase compliance is the so called ”But you

are free to accept or refuse technique” that was tes-

ted by (Gu

eguen and Pascual, 2005). The technique

works by underlining the option to refuse a request. In

that speciﬁc study, Gu

eguen and Pascual asked sub-

jects on the streets whether they want to participate

in a survey. Those subjects that were asked using the

technique, that is by emphasizing the possibility to re-

fuse the request, showed a signiﬁcantly higher com-

pliance rate, compared to the control group in which

the option to refuse was not emphasized (Gu

eguen

and Pascual, 2005). The ”But you are free to accept or

refuse”-technique is a simple and easy-to-implement

way that can bring improvements especially in persu-

asive dialogue or recommendation situations.

Adding a bulky afﬁx to sentences is surely a rather

inelegant way of increasing compliance. The other

hand, the principle of emphasizing the option of refu-

sal can also be applied by other means. For example,

The Beneﬁts of Considering Psychological Reactance as Usersâ

Z Personality Trait in HCI Research - Proﬁling Users with Hongâ

Psychological Reactance Scale

161

in the case of a graphical user interface, by promi-

nently adding an abort or refuse button. For the seat

belt alarm, a button to shut down the warning is de-

fying the importance of the message and a successful

persuasion. However an option to postpone the alarm

for a couple of seconds could have a similar effect on

reactance because it gives some control to the driver.

Personalities that show a high level of trait reactance

should still be treated accordingly and as the study of

eguen and Pascual showed, afﬁxes to formulations

can have a signiﬁcant effect (Gu

eguen and Pascual,

2005). In order not to reverse the intended effect by

annoying the user, dialogues would have to contain a

larger body of phrases that emphasize the freedom of

refusal, which would also be needed to be harmoni-

ously inserted into the dialogue.

Commanding Tone. Commanding tone can be ad-

justed by the use of strong, commanding phrases like

”you must to” or ”you absolutely have to” versus ”you

could do” or ”why don’t you do”. Adapting the mag-

nitude of commanding tone in dialogues to trait re-

actance could be an especially promising approach to

adjust to users. In contrast to emphasizing the free-

dom of choice, a highly commanding tone empha-

sizes the freedom threat. This is the opposite effect

as the ”But you are free to accept or refuse techni-

que”. Using a commanding tone on the other hand

adds a component of authority to a message, there-

fore, it could be beneﬁcial for dialogue systems to em-

ploy commanding tone in dialogues that are meant to

persuade users. Again, the use of commanding tone

could be adjusted to the level of trait reactance of a

user in order to achieve the most effective persuasion.

Hereby, users with a low level of trait reactance can

be exposed to a higher amount of commanding tone,

in order to increase persuasion or motivation, whereas

users with a high level of reactance should be exposed

to only minor levels of commanding tone in order not

to spoil persuasion or motivation by inducing reactant

behavior. For the seat belt alarm this would imply,

that drivers with low trait reactance could be addres-

sed more strongly whereas drivers with high trait re-

actance would have to be addressed more politely in

order to achieve the most effective persuasion.

4.2 Inducing Reactant Behavior

Just as there are many situations in which reactance is

better be avoided, there are some situations in which

reactance can be beneﬁcial. This are mostly situations

in which the user is meant to be motivated. One sort

of applications that would beneﬁt from such effects

are ﬁtness applications.

Target-directed Reactance. In ﬁtness applications,

reactance could be induced by giving the user the im-

pression that he or she is underestimated by the appli-

cation. This can be done by presenting goals that the

user can accomplish, like running a certain distance

in 30 minutes, but also indicate that the user is not

yet ready for those goals. The way that this is done

would have to be adjusted to the individual user’s trait

reactance. Highly reactant personalities are probably

already sufﬁciently reactant when the particular goal

is indicated as a goal for the future: ”Only two trai-

ning lessons until you can run three miles in 30 minu-

tes.”, whereas users with a low level of trait reactance

might require stronger stimuli: ”According to your

schedule, you are not yet allowed to run three miles

in 30 minutes.” This direct stimulation of reactance is

at risk of bearing negative implications for the user’s

opinion towards the ﬁtness application itself. This is

because the ﬁtness application is the direct source of

the threat and reactance behavior will be therefore di-

rected towards it. It is also important not to use mes-

sages that might have negative effects on the users’

self-esteem for inducing reactance, since this can add

to depression or neurosis.

Redirecting Reactance. For a ﬁtness application

(and any device or service) it should be avoided to be-

come the target of reactant behavior. In order to still

be able to induce reactance in its users, the ﬁtness ap-

plication would have to present another target towards

that the reactance behavior can then be directed. One

such target that is proposed here, is the average user

of that ﬁtness application. Alongside the personal go-

als and progress, additional information could be pre-

sented that shows the average progress of other users.

This information is intended to induce reactance to-

wards the average user, therefore the achievements of

the average user have to be better, compared to the

achievements of the actual user. Since the actual user

will often be above average, the application would

have to either fake the average achievements, or use

only the achievements of the most-performing users,

e.g. the 50th percentile. Strength of the threat can be

adjusted to the user’s level of trait reactance by decre-

asing the percentile. For a highly reactant user it is

probably sufﬁcient to present a large percentile, whe-

reas a user with a low level of trait reactance might re-

quire a small percentile. An example would be: ”For

a three mile run, you take 30 minutes, this is ﬁve mi-

nutes slower than the average performance of the top

30% of the users.”

HUCAPP 2017 - International Conference on Human Computer Interaction Theory and Applications

162

4.3 Findings in Literature

The concept of reactance has already been addressed

to some extend by the HCI community.

For example, Roubroeks et al. found, that persu-

asive attempts of an online recommendation system

induced more reactance when they were accompanied

by socially relevant agents (Roubroeks et al., 2009;

Roubroeks et al., 2011). They used still images and

videos of a virtual robot as a social agent. On the ot-

her hand, Choi et al. found evidence that the level

of social presence of a virtual agent can positively in-

ﬂuence user’s trust in the stimulus (Choi et al., 2001),

therefore, a socially relevant agent might be beneﬁcial

for persuasive systems in some aspects. The ﬁndings

of Roubroeks et al. and Choi et al. imply that so-

cial presence of virtual agents or IPAs could improve

trust in services but that this comes with the risk of

inducing state reactance. The knowledge of a persons

trait reactance could help to determine the optimal le-

vel of social presence that a service or an IPA should

emit for each user individually and thereby increase

acceptance by optimizing the level of trust(Choi et al.,

2001) and reactance (Roubroeks et al., 2009; Rou-

broeks et al., 2011).

The two-sidedness of reactance in persuasive sy-

stems was also noted by Lee and Lee (Lee and Lee,

2009) They found that even though adaption of the

recommendation system to the user can improve the

user’s intention to use the service again, this effect can

tip over at some point and induce reactant behavior,

such as the intention to avoid using the service. De-

termining the individual trait reactance of users could

probably help to determine their individual tipping

point with a user model and improve the adaptation

strategies of recommendation systems.

5 CONCLUSION AND FUTURE

RESEARCH

A user’s level of trait reactance is a valuable informa-

tion to have for developers. Such knowledge can be

used for a couple of purposes.

It can be used to optimize a trade-off between per-

suasive effort and compliance in a situation in which

users are meant to be persuaded or receive recommen-

dations. For example, an artiﬁcial ﬁtness coach could

use such information to adjust its motivation beha-

vior to the user. A user that has a low level of trait

reactance could be exposed to stronger motivational

cues more frequently. A user who has a high level of

trait reactance is annoyed more easily and could then

abandon training or switch to another ﬁtness coach.

Such a user might receive a smaller number or less

intrusive motivational cues.

The concept of psychological reactance has alre-

ady been recognized by the marketing industry and

also received attention in the respective research com-

munity (Dillard and Shen, 2005; Tucker, 2014; Ed-

wards et al., 2002). As sophisticated services, such

as IPAs are getting more human-like and their percei-

ved social relevance increases, they are getting more

prone to induce effects like reactance. This was also

shown by Ehrenbrink et al., who showed that an adap-

tive spoken dialogue system can induce more reac-

tance, compared to a non-adaptive spoken dialogue

system (Ehrenbrink et al., 2016b).

To date, utilizing trait reactance is possible and

could prove beneﬁcial for human-computer inte-

raction. The utility of trait reactance is also due

to the fact that it does not need to be assessed fre-

quently and will still be valid month after the mea-

surement. As a future perspective, also the utiliza-

tion of state reactance is an option. State reactance

would allow for higher adaptation precision, because

environmental factors that inﬂuence state reactance

would also implicitly be adapted for. The continu-

ous or frequent assessment of state reactance is a pro-

blem, though. Today, state reactance is often asses-

sed via a combination of a questionnaire and thought

listing (Dillard and Shen, 2005). This technique is

however unpractical for continuous measurement be-

cause users will hardly be willing to undergo such a

procedure frequently while interacting with some ser-

vice. Therefore, in order to be able to utilize state

reactance, new assessment techniques need to be de-

veloped. Such techniques could for example employ

behavioral or physiological data from wearable devi-

ces to draw conclusions on state reactance.

ACKNOWLEDGEMENTS

This work was supported by BMBF, Software Cam-

pus, grand no. 01IS12056, Sozialpsychologische As-

pekte von Smart Homes (SPASH).

REFERENCES

Apple Inc. (2011). Siri. Retrieved December 23, 2016 from

http://www.apple.com/ios/siri/.

Brehm, J. W. (1966). A Theory of Psychological Reactance.

Academic Press, New York.

Brehm, S. S. and Brehm, J. W. (1981). Psychological re-

actance: A theory of freedom and control. Academic

Press New York.

The Beneﬁts of Considering Psychological Reactance as Usersâ

Z Personality Trait in HCI Research - Proﬁling Users with Hongâ

Psychological Reactance Scale

163

Choi, Y. K., Miracle, G. E., and Biocca, F. (2001). The

effects of anthropomorphic agents on advertising ef-

fectiveness and the mediating role of presence. Jour-

nal of Interactive Advertising, 2(1):19–32.

Dillard, J. P. and Shen, L. (2005). On the nature of reac-

tance and its role in persuasive health communication.

Communication Monographs, 72(2).

Edwards, S. M., Li, H., and Lee, J.-H. (2002). Forced ex-

posure and psychological reactance: Antecedents and

consequences of the perceived intrusiveness of pop-up

ads. Journal of Advertising, 31(3):83–95.

Ehrenbrink, P., Gong, X. G., and M

oller, S. (2016a). Im-

plications of different feedback types on error percep-

tion and psychological reactance. In Proceedings of

the Annual Meeting of the Australian Special Interest

Group for Computer Human Interaction. ACM.

Ehrenbrink, P., Hillmann, S., Weiss, B., and M

oller, S.

(2016b). Psychological reactance in hci - a method

towards improving acceptance of devices and servi-

ces. In Proceedings of the Annual Meeting of the Au-

stralian Special Interest Group for Computer Human

Interaction. ACM.

Goldberg, L. R. (1993). The structure of phenotypic perso-

nality traits. American psychologist, 48(1):26.

Google Inc. (2012). Google now. Retrieved December 23,

2016 from https://www.google.com/landing/now/.

eguen, N. and Pascual, A. (2005). Improving the re-

sponse rate to a street survey: an evaluation of the”

but you are free to accept or to refuse” technique. The

Psychological Record, 55(2):297.

Hassenzahl, M. and Monk, A. (2010). The inference of

perceived usability from beauty. Human–Computer

Interaction, 25(3):235–260.

Hong, S.-M. (1992). Hong’s psychological reactance scale:

a further factor analytic validation. Psychological Re-

ports, 70(2):512–514.

Hong, S.-M. and Faedda, S. (1996). Reﬁnement of the hong

psychological reactance scale. Educational and Psy-

chological Measurement, 56(1):173–182.

Hong, S.-M. and Page, S. (1989). A psychological reac-

tance scale: Development, factor structure and relia-

bility. Psychological Reports, 64(3c):1323–1326.

ISO 9241-11 (1999). Ergonomische anforderungen fr brot-

tigkeiten mit bildschirmgerten teil 11: Anforderungen

an die gebrauchstauglichkeit - leitstze.

Jonason, P. K. and Knowles, H. M. (2006). A unidimensio-

nal measure of hong’s psychological reactance scale.

Psychological Reports, 98(2):569–579.

Lee, G. and Lee, W. J. (2009). Information & manage-

ment psychological reactance to online recommenda-

tion services. Information & Management, 46(8):448–

452.

Likert, R. (1932). A technique for the measurement of atti-

tudes. Archives of Psychology.

Looije, R., Neerincx, M. A., and Cnossen, F. (2010). Per-

suasive robotic assistant for health self-management

of older adults: Design and evaluation of social be-

haviors. International Journal of Human-Computer

Studies, 68(6):386–397.

Merz, J. (1983). Fragebogen zur messung der psychologis-

chen reaktanz. Diagnostica.

Microsoft (2014). Cortana. Retrieved December 23, 2016

from https://www.microsoft.com/windows/cortana/.

Rains, S. A. (2013). The nature of psychological reactance

revisited: A meta-analytic review. Human Communi-

cation Research, 39(1):47–73.

Roubroeks, M., Ham, J., and Midden, C. (2011). When

artiﬁcial social agents try to persuade people: The role

of social agency on the occurrence of psychological

reactance. International Journal of Social Robotics,

3(2):155–165.

Roubroeks, M., Midden, C., and Ham, J. (2009). Does it

make a difference who tells you what to do ? explo-

ring the effect of social agency on psychological reac-

tance. In Persuasive. ACM.

Shen, L. and Dillard, J. P. (2010). Psychometric properties

of the hong psychological reactance scale. Jornal of

Personality Assessment, 85(1):74–81.

Silvia, P. J. (2005). Deﬂecting reactance: The role of simila-

rity in increasing compliance and reducing resistance.

Basic and Applied Social Psychology, 27(3):277–284.

Tucker, C. E. (2014). Social networks, personalized ad-

vertising, and privacy controls. Journal of Marketing

Research, 51(5):546–562.

White, T. B., Zahay, D. L., Thorbjørnsen, H., and Shavitt,

S. (2008). Getting too personal: Reactance to highly

personalized email solicitations. Marketing Letters,

19(1):39–50.

HUCAPP 2017 - International Conference on Human Computer Interaction Theory and Applications

164