=Paper=
{{Paper
|id=None
|storemode=property
|title=Human Factors Limiting Consumer Benefit from Decisional Support
|pdfUrl=https://ceur-ws.org/Vol-973/bcss2.pdf
|volume=Vol-973
|dblpUrl=https://dblp.org/rec/conf/persuasive/PhillipsOB13
}}
==Human Factors Limiting Consumer Benefit from Decisional Support==
https://ceur-ws.org/Vol-973/bcss2.pdf
First International Workshop on Behavior Change Support Systems (BCSS 2013) 15
HumanFactorsLimitingConsumerBenefitfromDeǦ
cisionalSupport
James Phillips1, Rowan Ogeil2, and Alex Blaszczynski3
1
Monash University, Clayton, Australia
jim.phillips@monash.edu
2
Turning Point Alcohol and Drug Centre, Fitzroy, Australia
rowano@turningpoint.org.au
3
University of Sydney, Sydney, Australia
alex.blaszczynski@sydney.edu.au
Abstract. We cite examples from our own research indicating that those
populations requiring decisional support to change specific aspects of their
behavior may actually be the least able to benefit from that decisional sup-
port. Implications for designers are discussed.
Keywords: decisional support, behavior change, gambling, aggregation
1 Overview
Although the Internet allows providers to deliver advice and assistive services 24
hours, 7 days a week [1], this enhanced availability may not equate with an enhanced
user receptivity to such assistance. Our studies indicate that the availability of deci-
sional support may not equate with the effective use of this decisional support. These
studies have implications for the design of behavior change support systems by high-
lighting the potential need for systems to “pull” or personally contact individual users,
or to aggregate their services over multiple providers.
2 Procrastination
For a Behavior Change Support System [1] to be used, it needs to be accessed or sub-
scribed to, and some individuals may not yet be contemplating or be committed to a
change. Procrastinators are individuals who tend to postpone actions or decisions.
Procrastination is associated with mental health problems such as depression [2],
Although procrastinators are likely to require support modifying their behavior [3],
there are reasons to suspect that they may be less able to use that help. We have
tracked users of Learning Content Management Systems, and correlated usage with
self-reported decisional style [4], finding that procrastinators were less likely to actu-
ally login! Hence support systems [3] should probably chase such users rather than
hope users will access the system. Such observations argue for behavior change sys-
tems to contact users when the system has been inactive preferably using a more per-
sonal device such as a mobile phone.
However, such “nagging” technologies may be less effective for procrastinators
even if they are actually logged in. Within work environments, Shirren and Phillips
16 First International Workshop on Behavior Change Support Systems (BCSS 2013)
[5] used a diary study to examine responses to emails. Higher levels of email traffic
were associated with increased stress levels, and although procrastinators reported
reading messages, they were slower responding to messages. Indeed, other studies
have suggested that access to specific learning support systems may be reduced when
people have other tasks to perform [6]. Hence designers need to appreciate that some
users may be harder to help than others and acknowledge that system access and use
may vary as a function of other commitments.
3 Mentation
Compliance, or changes in attitude or behavior [1] all necessitate some degree of user
response, hence there is a need to acknowledge user status. When people are men-
tally impaired they are actually in greater need of assistance. We have conducted a
series of studies systematically examining people’s ability to avail themselves of deci-
sional support. We employed a simulated gambling task and provided a variety of
forms of online advice that would either guide participant’s strategies or inform par-
ticipants as to when the odds were in their favour. In such simulated gambling stud-
ies, participants complying with this decisional support can minimise their losses.
So far, we have considered response to decisional support as a function of alcohol
intoxication [7] and sleep deprivation [8]. The general finding was that people re-
quired more time to make use of advice, and were least able to use the decisional
support when they needed it the most, namely when they were intoxicated or sleep
deprived, when there was time pressure, higher risk, or when they were losing. On a
more positive note, people with impaired mentation could use and indeed were more
reliant upon decisional support, but there remains a need to acknowledge that there
were cognitive costs associated with decisional support.
Such studies indicate that any improvement associated with decisional support is
not “free”. It will require user time and effort to process. Indeed in a study of “brain
training” we found that any benefits on cognitive function were associated with
greater dedication and use of the “brain training” software [9].
4 Multiple Providers
Behavior Change Support Systems [1] targeting specific behaviors such as substance
abuse or gambling need to be aware of the difficulties posed by multiple providers.
Although it is possible to consider the context under which the user operates [10],
there are behaviors that are of interest to regulators, that off-shore providers and end-
users may not be quite so interested in curbing. There are also political and financial
problems and issues of privacy to be resolved when developing systems to help an
individual block themselves from gambling or substance abuse.
We have found that problem drinkers and problem gamblers frequent more ven-
ues, and that an intervention targeting a specific location (i.e., self-barring) may not
be effective if a person also frequents other locations [11]. Indeed prescription sub-
stance abusers also appear to frequent more sources of drugs, shopping from doctor to
doctor [12-13]. More importantly we have also observed that problem gamblers ac-
cess a wider range of gambling products [14], suggesting that reducing the number of
First International Workshop on Behavior Change Support Systems (BCSS 2013) 17
electronic gambling machines or blocking access to a specific operator is unlikely to
be effective if gamblers can access gambling by other means (e.g., internet or mobile
phones). The same problems occur with controlled substances [15,16] and to combat
this, central registers and a list of behaviors predictive of doctor shopping have been
employed [12]. Systems seeking to support a client’s need to reduce gambling, or
their access to drugs [12,17] need go beyond targeting one specific source of supply.
5 Implications
Although it is assumed that an increased number of persuasive messages will lead to
greater compliance, this is not guaranteed. For instance a study of interpersonal influ-
ence [18] found curvilinear relationships between the degree of influence and re-
peated messages from the same source. With repeated attempts to influence, habitua-
tion occurs and impacts lessen.
As our ongoing studies demonstrate, a Behavior Change Support System [1] needs
to be attended to, and available capacity may be a function of workloads. In particu-
lar, people may be least able to process decisional support when it is needed the most.
Hence systems developed and tested upon normally functioning people may not be
optimal for impaired populations. These findings should not be surprising, as our
earlier studies on the use of meaningful contingent cues typically found greater de-
pendence upon cues and a diminished ability to use cues in impaired populations.
Indeed from experience with a wide range of clinical populations (Alzheimer’s, Hunt-
ington’s, Parkinson’s disease, Schizophrenia, and Major depression) those individuals
requiring cues to assist with a specific deficit are actually least able to utilize those
cues, indeed this is why they have problems and require assistance [19]. There have
been some reports that tonic non-contingent cues may serve to “pace” the speed or
frequency of behavior, but there have been problems replicating these findings. On-
going studies have noted the failure to benefit from cues under such circumstances
where people’s performance is paced and under time pressure [8].
6 Conclusions
Behavior Change Support Systems [1] require effort on the part of the user. Indeed
those individuals requiring support may be least able to avail themselves of this sup-
port. Support systems may need to take workloads into account, but actively contact
users during periods of inactivity. In addition, consumer mobility argues that support
systems may sometimes need to take an overarching role of aggregator and deal with
multiple service providers.
References
1. Oinas-Kukkonen, H.: Behavior Change Support Systems: A research model and
agenda. Lect. Notes. Comp. Sci. 6137, 4-14 (2010)
2. Radford, M.H.B., Mann, L., Kalucy, R.S.: Psychiatric disturbance and decision
making. Aus. N.Z. J. Psychiat. 20, 210-217 (1986)
18 First International Workshop on Behavior Change Support Systems (BCSS 2013)
3. Bernold, L.E.: Early warning system to identify poor time management habits.
Int. J. Eng. Educ. 23(6), 1182-1191 (2007)
4. Phillips, J.G., Jory, M., Mogford, N.: Decisional style and eParticipation. Pro-
ceedings of the 19th Australasian conference on Computer-Human Interaction:
Entertaining User Interfaces, pp. 139-141 (2007)
5. Shirren, S., Phillips, J.G.: Decisional style, mood and work communication:
email diaries. Ergonomics, 54, 891-903 (2011)
6. Phillips, J.G., Hyman, G., Bertling, P, Jory, M.: Videoconferenced lectures are a
preferred download. Proceedings of the 12th Annual Conference of the New Zea-
land Chapter of the ACM Special Interest Group on Computer-Human Interac-
tion, pp. 113-116 (2011)
7. Phillips, J.G., Ogeil, R.P.: Alcohol influences the use of decisional support. Psy-
chopharmacol. 208, 603-611 (2010)
8. Fraser, M., Conduit, R., Phillips, J.G.: Effects of sleep deprivation upon deci-
sional support utilisation. Ergonomics 56(2), 235-245 (2013)
9. James, L., Phillips, J.G., Best, C.: Effect of “Brain Training” and gaming on fluid
intelligence. J. CyberTher. Rehab. 4, 433-446 (2011)
10. Dourish, P.: What we talk about when we talk about context. Pers. Ubiq. Com-
put. 8(1), 19-30 ( 2004)
11. Phillips, J.G., Ogeil, R.P.: Decisional styles and risk of problem drinking or gam-
bling. Pers. Individ. Diff. 51 , 521-526 (2011)
12. Martyres, R., Clode, D., Burns, J.M.: Seeking drugs or seeking help? Escalating
“doctor shopping” by young heroin users before fatal overdose. Med. J. Aus. 180,
211-214 (2004)
13. Rigg, K. K., Kurtz, S. P., Surratt, S. L. Patterns of prescription medication diver-
sion among drug dealers. Drugs 19(2), 144-155 (2012)
14. Phillips, JG., Ogeil, R., Chow, Y-W., Blaszczynski, A.: Gambling involvement
and increased risk of gambling problems. J. Gamb. Stud. DOI 10.1007/s10899-
012-9325-x
15. Ivanitskaya, L., Brookins-Fisher, J., O’Boyle, I., Vibbert, D., Erofeev, D., Fulton,
L.: Dirt cheap and without prescription: How susceptible are young US consum-
ers to purchasing drugs from rogue Internet pharmacies. Jour. Med. Internet Res,
12(2) (2010)
16. Nielsen, S., & Barratt, M. J.: Prescription drug misuse: Is technology friend or
foe? Drug Alcohol Rev., 28, 81–86 (2009)
17. Nower, L., Blaszczynski, A.: Gambling motivations, money limiting strategies,
and precommitment preferences of problem versus non-problem gamblers. J.
Gamb. Stud. 26, 361-372 (2010)
18. Leskovec, J., Adamic, L.A., Huberman, B.A.: The dynamics of viral marketing.
ACM Trans. Web. 1(1), 1-39 (2007)
19. Phillips, J.G., Stelmach, G.E. Parkinson's disease and other involuntary move-
ment disorders of the basal ganglia. In: Fredericks, C. & Saladin, L. (eds.) The
motor system: basic principles and pathophysiology. F.A. Davis, Philadelphia,
pp. 424-444 (1996)