Gait & Posture 37 (2013) 470–471

Contents lists available at SciVerse ScienceDirect

Gait & Posture

journal homepage: www.elsevier.com/locate/gaitpost

Short communication

Nonconscious priming of elderly or youthful stereotypes influences walking

velocity in healthy undergraduates
Olga Stanislavova a, Steven J. Robbins a,*, Carol A. Oatis b
a
Department of Psychology, Arcadia University, 450 South Easton Road, Glenside, PA 19038, United States
b
Department of Physical Therapy, Arcadia University, 450 South Easton Road, Glenside, PA 19038, United States

A R T I C L E I N F O A B S T R A C T

Article history: Psychological studies of ‘‘priming’’ effects document that covertly presented stimuli can influence
Received 23 September 2011 behavior without participant awareness. We examined whether nonconscious priming can influence
Received in revised form 1 June 2012 walking velocity on an instrumented walkway. Fifty-nine healthy participants were randomly assigned
Accepted 30 August 2012
to unscramble sentences and answer self-concept questions containing words related to either the
concepts elderly/passive or youthful/active. Walking velocity was measured before and after this priming
Keywords: manipulation. An ANOVA revealed a significant Group  Time interaction; the elderly/passive group
Gait
showed a decrease in walking velocity over time while the youthful/active group showed an increase.
Walking velocity
Priming
None of the participants expressed any awareness of the study purpose and the experimenter was blind
Nonconscious to group assignment. These results suggest that spoken or written material received by patients prior to
gait assessments (such as casual conversations or written health inventories) might influence walking
performance outside of client or clinician awareness.
ß 2012 Elsevier B.V. All rights reserved.

1. Introduction 2. Methods

Recent studies of ‘‘priming effects’’ document the powerful 2.1. Participants

impact of nonconscious mental events on behavior [1–4]. For
example, Bargh et al. asked a group of participants to unscramble
sentences including words related to the concept elderly; neutral
words replaced the elderly words in a control group [2]. Individuals
exposed to elderly concept words walked more slowly down a
hallway compared to controls, even though neither group was
aware of the purpose of the priming stimuli (or that walking was
being measured). Priming effects might play a significant and
undetected role in determining performance in clinical settings as
well. Both conversations and self-report assessments which take
place at the outset of an appointment might activate mental
categories such as disability, pain, or activity.
To investigate this possibility, we randomly assigned healthy
participants to written materials priming the concepts elderly/
passive or youthful/active. Walking velocity was measured before
and after priming using a standard instrumented walkway. We
expected the elderly/passive primes to reduce walking speed and
the youthful/active primes to produce an increase. We further
expected that individuals in neither group would be able to
articulate the purpose of the study during debriefing.
* Corresponding author. Tel.: +1 215 572 2987.
E-mail address: robbins@arcadia.edu (S.J. Robbins).
Participants were 59 healthy, undergraduate volunteers at
Arcadia University between the ages of 18 and 29 (mean age = 21.1,
SD = 2.3). All participants were recruited either through adver-
tisements posted on campus or individually by the first author.
2.2. Materials
Walking performance was measured using a GaitMat IITM
instrumented walkway 3.87 m long, 0.81 m wide, and 0.03 m deep.
Foot contact on the mat closed pressure-sensitive switches and
contacts were recorded digitally on an interfaced computer. We
measured walking velocity (m/s) based on the average of all steps
taken on the walkway. Previous studies have verified the reliability
of GaitMat II measurements [5,6].
2.3. Procedure
Following written consent, the first author read to each
participant a cover story claiming that the study was examining
the effects of exercise on cognitive performance (reversing the true
cause and effect elements of the study). Participants were told that
the ‘‘exercise manipulation’’ required them to cross the walking
mat at one of several speeds (normal walking, jogging, running)

0966-6362/$ – see front matter ß 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.gaitpost.2012.08.018
 O. Stanislavova et al. / Gait & Posture 37 (2013) 470–471
Table 1
Mean (SD) walking velocity (m/s) on the third trial of the pretest and posttest
periods by group.
Group N Pre Post Pre–post
Elderly/passive 29 1.39 (0.23) 1.36 (0.22) +0.03 (0.10)
Youthful/active 30 1.32 (0.13) 1.35 (0.20) 0.03 (0.11)
prior to completing a ‘‘cognitive task.’’ In actuality, the ‘‘cognitive
task’’ served as the priming manipulation and all individuals were
asked to walk across the mat at ‘‘your normal walking speed’’. The
cover story was intended to prevent awareness of the priming
manipulation and that walking velocity was in fact the outcome
measurement.
At baseline, each participant walked across the mat three times.
On each trial, the individual began walking at least 1 m in advance
of the walkway and for at least 1 m beyond the end of the mat. We
analyzed walking velocity on the final trial. Participants were then
given the priming stimuli in sealed envelopes to ensure the
experimenter (first author) remained blind to group assignment.
Half the participants were randomly assigned to receive materials
priming the concepts elderly/passive (n = 29); half received
youthful/active priming materials (n = 30).
The first priming task involved 30 sets of five words in
scrambled order; participants were asked to form sentences from 4
of the 5 words in each set. Word sets presented to the elderly/
passive group contained embedded elderly concept words (e.g.,
‘‘him was worried she always’’); the youthful/active group received
youthful concept words in the same word sets (‘‘him was carefree
she always’’). The priming task used by Bargh et al. [2] served as the
template for this manipulation.
The second priming task required individuals to fill out a
supposed ‘‘self-concept’’ questionnaire. Respondents rated seven
statements on a visual-analog agree/disagree scale. Statements
were phrased to ensure that individuals would generally express
some level of agreement with each item. In the elderly/passive
condition the sentences led individuals to self-label with passive
terms (‘‘Sometimes, others take care of me’’); in the youthful/active
condition the statements expressed active self-attributions (‘‘I take
care of myself’’).
The two written tasks were linked. The elderly/passive group
received the elderly sentence scrambles and the passive self-
concept items; the youthful/active group received the youthful
sentence scrambles and the active self-concept sentences.
Following priming, participants traversed the walkway three
times (posttest), again with instructions to ‘‘walk at your normal
speed.’’ During a funneled debriefing, none of the participants
expressed any awareness of the priming manipulation.
3. Results
Mean (SD) walking velocities at pretest and posttest for the two
groups during the third walking trial are depicted in Table 1. A 2-
factor, repeated measures ANOVA revealed a significant Group by
Time interaction (F(1, 57) = 5.2, p < 0.05). Changes in velocity were
471
in the expected directions, faster in the youthful/active condition
and slower in the elderly/passive condition. This result was not
specific to use of the third walking trial as the dependent measure;
a follow-up analysis using the average of all three walks also
revealed a significant interaction (F(1, 57) = 4.6, p < 0.05).
4. Discussion
Individuals primed with the categories elderly/passive showed a
reduction in walking velocity on the Gait Mat II walkway while
individuals primed with the concepts youthful/active subsequently
increased their speed. Two elements of the study deserve note.
First, random assignment to priming conditions permits the causal
conclusion that activated belief states influenced walking speed.
Second, this design rules out demand effects and observer bias as
confounds; the experimenter was blind to group assignment and
the cover story prevented participant awareness of the study
hypotheses. Because gait velocity is a common clinical outcome
measure, the implications for clinical practice are straightforward;
conversations and self-report inventories occurring at session
outset may prime changes in physical functioning without the
awareness of either therapist or client.
Inadvertent priming could affect behavior in other populations
as well. For example, a conversation about an elderly patient’s
grandchildren might prime a mental construct of young while a
conversation about their friends might prime the construct old.
Similarly, discussion of active pursuits such as taking a trip or
paying a visit might prime mental categories leading to faster
walking, whereas discussions of recent health problems or passive
activities (receiving a visit from children) might prime slower
walking. Further studies are necessary to pursue such questions
and to extend the present results to disabled and elderly
participant samples.
Conflict of interest statement
None of the authors have any conflicts of interest.
References
[1] Bargh JA, Chartrand T. Studying the mind in the middle: a practical guide to
priming and automaticity research. In: Reis H, Judd C, editors. Handbook of
research methods in social psychology. New York: Cambridge University Press;
2000. p. 1–40.
[2] Bargh JA, Chen M, Burrows L. Automaticity of social behavior: direct effects of
trait construct and stereotype activation on action. Journal of Personality and
Social Psychology 1996;71:230–44.
[3] Bargh JA, Gollwitzer PM, Lee-Chai A, Barndollar K, Trotschel R. The automated
will: nonconscious activation and pursuit of behavioral goals. Journal of Per-
sonality and Social Psychology 2001;81:1014–27.
[4] Ferguson MJ, Bargh JA. How social perception can automatically influence
behavior. Trends in Cognitive Sciences 2004;8:33–9.
[5] Barker S, Craik R, Freedman W, Herrman N, Hillstrom H. Accuracy, reliability,
and validity of a spatiotemporal gait analysis system. Medical Engineering and
Physics 2006;28:460–7.
[6] Leiper CI, Craik RL. Relationships between physical activity and temporal-
distance characteristics of walking in elderly women. Physical Therapy
1991;71:791–803.