PERSONAL FITNESS TRAINER: THE EFFECT OF FEEDBACK
PRESENTATION FORMATS
Duwaraka Yoganathan, Department of Information Systems, School of Computing, National
University of Singapore, Singapore, [email protected]
Sangaralingam Kajanan, Department of Information Systems, School of Computing, National
University of Singapore, Singapore, [email protected]
Abstract
This paper explores the effect of feedback presentation formats in mobile fitness apps. Drawing from
computers are social actors (CASA) paradigm, cognitive load theory (CLT), and social presence
literature in this study, we conceptualize that when a user is involved in different types of exercise
(aerobic or anaerobic) the exposure to more humanized feedback can create differential influence in
user’s cognitive load and user engagement. This differential influence subsequently affects the fitness
performance of the user. This conceptualization will be validated through an experiment. The
outcomes of this study can help fitness app developers to design effective fitness apps. Similarly the
findings can assist users in choosing the most appropriate fitness app that can enhance their fitness.
Keywords: Fitness apps, Feedback presentation, humanized feedback, Computers are social actors
(CASA), Cognitive load.
1
INTRODUCTION
Smartphone industry has experienced an exponential growth in the recent past. These smart devices
run third party software applications called mobile apps, which provide us with various functionalities
beyond just making calls. Mobile apps not only allow us to play games, check email, and watch
movies but also help us to lead a healthy lifestyle. These health and fitness apps have a great potential
to assist us in leading a healthy lifestyle as they are versatile and are being with us always at our
fingertips. Latest health and fitness apps are built-in with various functionalities such as GPS
technology (e.g. RunKeeper, MapMyRun), calorie calculator (e.g. MyFitnessPal, Fitbit) and heart rate
monitor (e.g. Instant Heart Rate) to assist users in grasping fine grained details of activity. At the same
time these apps employ various techniques such as exergames or gamification (e.g. Zombies Run,
Walkathon + Fitness Games), music (e.g. PaceDJ – Syn Your Running Pace With Your Music!),
social networking (e.g. Fitocracy), customization (e.g. mywellness) and reward system (e.g.
RunKeeper) to motivate greater physical activity performance. In addition, fitness apps support
various kinds of physical activity types such as running (e.g. RunKeeper, Edomondo, Map My Run),
jogging (e.g. iRunner ), walking (e.g. Map My Walk+), bicycling (e.g. Map My Ride, Wahoo Fitness),
weight lifting (e.g. Gym Training) and pushups (e.g. Hundred Pushups). As of February 2014 there are
more than 40,000 health and fitness apps available for consumers in major app stores1.
Despite the growing demand for mobile apps in healthcare, research related to mobile apps is limited
(Yoganathan & Kajanan, 2013). Therefore it is important to study mobile app design factors that can
enhance the health and wellbeing of individuals.
Prior literature on physical activity have revealed that self-monitoring (Lin et al., 2006; Tsai et al.,
2007; Consolvo, McDonald, et al., 2008; Mattila et al., 2008), music (Oliver & Flores-Mangas, 2006),
exergames or gamification (Fujiki et al., 2007), competition (Connelly et al., 2006; Lin et al., 2006),
social support (Gasser et al., 2006; Consolvo, Klasnja, et al., 2008; Ahtinen et al., 2009) , reminders
(Lee et al., 2006), sharing performance with friends(e.g. Houston (Consolvo et al., 2006)) and stylized
displays such as virtual pet (e.g. fish N’ steps (Lin et al., 2006)) and garden metaphor (Consolvo,
McDonald, et al., 2008) can enhance the physical activity behaviour of individuals. However, design
of feedback mechanism has not yet been examined.
Therefore in this study, we investigate on the design of feedback mechanism in fitness apps. Prior
research suggest that providing timely and appropriate feedback important as it can motivate
individuals and reinforce the exercise behaviour (Dubbert et al., 1984). Thus, the manner in which
feedback is presented to the user can greatly influence the exercise outcomes.
Studies related to human computer interaction indicate that humans react to computers as “social
actors” (Nass et al., 1994; Reeves & Nass, 1996; Nass & Moon, 2000). Therefore, feedbacks with
higher social presence (i.e. more social cues) can elicit greater user engagement as they would satisfy
human’s tendency to be social (Tung & Deng, 2006). However, for certain types of exercises (e.g.
strength training), a more humanized feedback would produce less performance and more fatigue
(Rube N, Secher NH, 1980; Rube & Secher, 1981). Therefore, this research aims to understand when
it is suitable to provide a more humanized (i.e. high social presence) feedback? And the underlying
mechanism through which social presence in fitness app feedback influences an individual’s exercise
behaviour. More precisely, our research question is “How and when social presence in fitness app
feedback enhances physical activity behaviour of users?” Drawing on computer are social actors
(CASA) paradigm (Nass et al., 1994; Reeves & Nass, 1996; Nass & Moon, 2000), cognitive load
theory (Sweller et al., 1998; Paas et al., 2003) and social presence literature (Short et al., 1976; Hess et
al., 2009; Horvath & Lombard, 2010), in this study we conceptualize that when a user is involved in
1
http://www.mobilewalla.com
different types of exercise, the exposure to more humanized feedback can create differential influence
in user’s cognitive load and user engagement. This differential influence subsequently affects the
fitness performance of the user. The outcomes of this study can help fitness app developers to design
effective fitness apps. Similarly the findings can also assist app users in choosing the most appropriate
fitness app that can enhance their fitness.
The structure of the paper is as follows. In the immediately following section, we explain the
theoretical background and research model. After which we provide the research design, followed by a
discussion of potential implication for research and theory.
2
THEORETICAL BACKGROUND AND RESEARCH MODEL
In order to understand how and when feedback presentation with social presence facilitates greater
exercise behaviour, we draw literature from computer are social actors (CASA) paradigm (Nass et al.,
1994; Reeves & Nass, 1996; Nass & Moon, 2000), cognitive load theory (Sweller et al., 1998; Paas et
al., 2003) and social presence literature (Short et al., 1976; Hess et al., 2009; Horvath & Lombard,
2010). We conceptualize that social presence influences physical activity behaviour through user
engagement (i.e. affective element) and cognitive load (i.e. user’s working memory). This influence is
moderated by the type of physical activity user is performing (see figure 1). The following sections
describe conceptual model in detail.
2.1
Social Presence
Short et al (Short et al., 1976) coined the term “social presence” to describe the degree of awareness of
the other person in the communication media (Short et al., 1976) i.e. degree to which the receiver
perceives the other person as a “real person”. Social presence is high, when the interaction is close to
face-to-face interaction or human interaction. Social presence has also been conceptualized as “user’s
feeling or perception of being connected with another intellectual entity” (Tu & McIsaac, 2002),
feeling of warmth, sociability, intimacy, immediacy (Nass & Moon, 2000) or perception of human
contact (Kumar & Benbasat, 2006; Hess et al., 2009). Thus, a user’s perception of social presence or
feeling of interacting with a social being can invoke social and emotional reactions such as high
involvement and motivation in the user (Lombard & Ditton, 1997; Witmer & Singer, 1998; Nass &
Moon, 2000).
Several prior studies reveal that human-computer interactions or relationships are basically social,
hence humans respond to computer as real people or “social actors”(i.e. elicit social behaviours such
as politeness while interacting with computers) (Nass et al., 1994; Nass & Moon, 2000; Kumar &
Benbasat, 2006). The computers are social actors (CASA) paradigm (Nass et al., 1994, 1995; Nass &
Moon, 2000) emphasize that people characterize computers as humans or social actors, when
computers display sufficient social cues. As smartphones are handheld computers, we define social
presence in fitness app feedback as “the degree to which a user perceives the feedback from fitness
app, as if a feedback received from a social actor”. The CASA literature provides four characteristics
for social actors (Nass et al., 1994, 1995; Reeves & Nass, 1996; Nass & Moon, 2000). The first
important characteristic is the language use. A language spoken in a natural manner similar to a human
is considered to be more social than other forms of languages(Winograd & Flores, 1986). For example
natural language style phrases such as “Come on guys!”, “Let’s do it!” can convey more social cues
than formal language styles. The second, characteristic is human sounding speech. An entity that
demonstrate a human sounding speech (i.e. human voice) will be perceived as a social actor, because
speech is human’s unique capability and human sounding speech is differently processed by human
brain than other sounds(Moore & Moore, 2003). The third characteristic is interactivity, i.e. the extent
to which the entity incorporates prior inputs (Rafaeli, 1990). Fourth characteristic is social roles,
computers that fulfil the social roles, traditionally filled by humans are perceived as social actors (e.g.
caregiver, doctor, coach), because individuals are defined or identified by the set of behaviours that are
associated with their roles (Kim et al., 2013).
As fitness app fulfil the role of fitness trainer and feedback is a response to user’s fitness behaviour
(interactive) out of these four characteristics we have chosen two characteristics (i.e. language use and
human sounding speech) to symbolize social presence in fitness app feedback.
Language use (informal/spoken style) and human sounding speech (voice) can convey important
social cues that exhibit social presence in fitness apps. E-commerce websites such as Amazon.com
have already started using spoken language style to convey social presence (Gefen & Straub, 2004)
(e.g., Amazon.com uses phrases like “Where’s My Stuff?”, “So You’d Like To….”, “Uh Oh! You
don’t have any….” in their website). Video games use informal language to motivate and engage users
(Barr et al., 2007; Turgut & Irgin, 2009). Studies related e-learning reveal that the use of more spoken
language has resulted in greater engagement among students than written language (Moreno & Mayer,
2000a, 2000b). Thus, we postulate that fitness app using informal spoken style language can bring in
enhanced exercise performance in users. For example an informal style feedback such as “Dude!!, you
have almost reached there!!, just 1 more mile to go” is more persuading than a formal style feedback
such as “Workout summary: Distance remaining: 1 mile..”. Similarly, a human sounding voice
feedback can deliver more nonverbal cues such as modulation, pitch, tone, and pauses that cannot be
communicated via visual format (Short et al., 1976). As human sounding voice is more natural and
familiar to users, it can elicit greater attachment and involvement. Therefore, greater exercise
performance can be anticipated.
2.2
User Engagement
Engagement, (according to the Merriam Webster dictionary) refers to “emotional involvement” or “a
state of being in gear”. Engaged users generally invest more time, attention, emotional attachment and
elicit desirable responses to computer interactions (Lehmann et al., 2012). O'Brien & Toms (O’Brien
& Toms, 2008) conceptualized user engagement, based on user experience attributes such as positive
affect, sensory appeal, endurability and interactivity. Moreover, studies related to web and mobile
applications, refer engagement as intensified frequent usage of applications(Claussen et al., 2012;
Lehmann et al., 2012). This study defines user engagement with fitness app as “a quality of experience
that captivates, attracts and motivates users in using the fitness app with interest”. Technologies aim to
humanize computer interactions, as more humanized interactions would be perceived as comfortable
and easy to use by users (Sproull et al., 1996). CASA paradigm indicates that, a human-computer
interaction that replicates or imitates human-human interaction with more social cues and intimacy
would enable individuals to mindlessly apply social rules to computers (Nass & Moon, 2000). This
kind of human-computer interaction would create feelings of warmth and sociability (Nass & Moon,
2000; Hess et al., 2009). According to social presence theory(Short et al., 1976), mere perception of
non-mediation or social presence aspects such as intimacy, immediacy and social cues can stimulate
feelings of anxiety, arousal, enjoyment and evoke positive social and emotional reactions in
individuals (Horvath & Lombard, 2010). Therefore, users would exhibit positive emotions towards
fitness apps when they perceive the feedback is from a social actor (i.e., when fitness app feedback
exhibits higher social presence). These positive emotions triggered through social presence can
enhance the user’s engagement towards the fitness app. Based on CASA paradigm and social
presence theory we argue that if users perceive higher social presence in fitness app feedback, they are
likely to be more involved with the app and would feel highly enthusiastic about using the app. Hence,
we hypothesize
H1: Higher social presence in fitness app feedback will result in higher user engagement in using the
app.
Studies related to fitness intervention reveal that users’ engagement with fitness interventions (i.e.
website delivered, internet and email based) significantly influenced the reduction of body weight
(Tate et al., 2001), increased exercise duration(Napolitano et al., 2003) and positive progress in
physical activity behaviour (Marcus et al., 1992, 1998). Therefore, we postulate that high engagement
with fitness app could yield positive physical activity outcomes. While users are enthusiastically
engaged with the fitness app, it is highly likely that they ignore the barriers, tiredness and difficulties
associated with physical activity behaviour. In addition, the ‘flow’ experience (Csikszentmihalyi,
1975; Csikszentmihalyi & Rathunde, 1993) and involvement(Lombard & Ditton, 1997) triggered by
the user engagement could stir the intrinsic motivation(Chapman, 1997) to engage in physical activity.
Therefore, users would enthusiastically participate in fitness activities. The “positive feeling states”
(Vealey, 2005; Hanin, 2007) or positive emotions (Vallerand & Blanchard, 2000) activated due to user
engagement could maximize the physical activity performance. Therefore, we expect that higher user
engagement with the fitness app will enhance the physical activity behaviour of users. Hence, we
hypothesize that
H2: Higher user engagement with fitness app will lead to better physical activity behaviour of users.
2.3
Cognitive load
Physical activity involves several cognitive processing functions. This includes not only executive
control processes such as coordination and planning but also effective processing of feedback
provided during exercise and improving the exercise performance accordingly. The cognitive load
theory (CLT) (Sweller, 1994; Sweller et al., 1998; Paas et al., 2003), defines cognitive load as the
“total amount of mental activity executed by the working memory at a given time and this working
memory is discrete and limited in capacity”. Cognitive load can have three distinctive parts: intrinsic
load, germane load and extraneous load (Paas et al., 2003). Intrinsic load refers to the inherent
complexity of the subject matter (Sweller et al., 1998). For example, a fitness workout that has simple
steps (e.g. jogging, running) has low intrinsic load than a workout that has number of complex steps
(e.g. yoga exercises) that require several elements to be handled simultaneously in working memory.
Germane load refers to the effort exerted in processing new information and integrating it into existing
knowledge structures (Paas et al., 2003). Thus, familiar exercises may be performed automatically
without high levels of conscious effort. This phenomenon is referred to as automaticity (Clark &
Mayer, 2008). Extraneous load is imposed by presentation format, which is also referred as ineffective
cognitive load (Paas et al., 2003). While intrinsic load is often integral and invariant to the subject
matter (Sweller, 1994), fitness app designers should target reducing the extraneous cognitive load in
order to allow greater amount of mental resources to be allocated to other relevant cognitive processes
that might be necessary to enhance the exercise motivation and efficiency.
Prior research suggest that extraneous cognitive load can be reduced by the effective design of
presentation format (Paas et al., 2003). For example, voice presentation will usually have lower
extraneous cognitive load than visual format, as voice would be processed separately by auditory
channel without overloading the working memory. Besides, a humanized voice is more comfortable
and familiar to individuals than a visual format, therefore less cognitive resources are required to
process voice than visual information (Chalfonte et al., 1991; Roda, 2011). Similarly, prior studies
indicate that information, presented in a familiar style, similar to a normal conversation would
consume less cognitive processing effort than processing a formal style information (Mayer, 1984;
Moreno & Mayer, 2000b; Roda, 2011). Individuals are more familiar and comfortable with the
informal spoken language style than formal language style (Walther, 1992; Nass & Steuer, 1993).
Hence, the cognitive load in processing the informal spoken style is lower compared to the formal
written style. Since informal spoken style is more humanized, it is likely to elicit greater social
presence. Therefore, a more humanized fitness app feedback with higher social presence would
consume less extraneous cognitive load. Hence, we hypothesize that
H3: Higher social presence in fitness app feedback will result in decreased cognitive load on working
memory.
As indicated above cognition is an important factor during physical activity. Specific cognitive
processes such as attention (e.g. awareness), planning (e.g. exercise steps, route), and decision making
(e.g. avoiding accidents) activities combinedly ensure safe and enduring physical activity behaviour.
For example, an effective brisk walk along the street/pathway, requires more cognitive effort and
attentional resources to be allocated for the activity itself (Lindenberger et al., 2000). While cognition
is essential to maintain a steady physical activity behaviour, cognitive overload can result in poor
exercise behaviour. Previous studies have shown that cognitive overload could results in ineffective
physical activity performance, such as taking longer than required time to complete physical activity
targets (Neider et al., 2010), reduced gait speed(Verghese et al., 2002), reduced motor skills (Verghese
et al., 2002), collisions and falls (Nagamatsu et al., 2011). Thus cognitive overload can jeopardize the
safely of users and cause injuries during physical activity. Therefore, we argue that cognitive overload
will negatively affect physical activity performance. Hence, we hypothesize that
H4: Cognitive overload on working memory will lead to poor physical activity behaviour.
2.4
Physical activity types
Physical activities can be categorized broadly into two main types as either aerobic or anaerobic based
on the energy source, by products, intensity of muscular contraction and duration of the exercise (Pate
et al., 1995; Haskell et al., 2007). Aerobic exercises means “with air”, in which our body utilizes
oxygen to create energy for the exercise and respiratory by-products (i.e. waste products) such as CO2
and water are easily cleared from body through breathing (Pryor & Kraines, 1999). Aerobic exercises
include activities such as walking, running, swimming, and bicycling. Health experts recommend a
minimum 30 minutes of daily aerobic exercises (Pate et al., 1995; Haskell et al., 2007). These aerobic
activities help in weight management, reduce the risk of lifestyle diseases (e.g. heart disease, type 2
diabetes) and improve physical fitness. In contrast, anaerobic exercise means “without oxygen”, in
which our body creates energy without oxygen (by breaking down sugar). This anabolic metabolism
produces lactic acid, which accumulates in our system and cause extreme fatigue and sometimes acute
pain in muscles (Medbo et al., 1988). Anaerobic exercises include strength trainings such as weight
lifting, gymnastics, body weight exercises, high intensity interval training and calisthenics such as
push-ups and pull-ups. Compared to aerobic exercises, anaerobic exercises are high intensity activities
that can last from 15 seconds to 2 minutes (Medbo et al., 1988). Health experts recommend 8-10
exercises in 2-3 times per week in order to build stronger muscles and reduced body fat within a short
period. While individuals are involved in aerobic and anaerobic exercises, the exposure to a
humanized feedback can create differential influence.
The studies conducted on anaerobic exercise (e.g. strength training) reveal that verbal encouragement
has resulted in reduced exercise performance and greater pronounced fatigue compared to those who
have not received verbal encouragement (Rube N, Secher NH, 1980; Rube & Secher, 1981).
Researchers offer reasons as to why this happens. While, an individual is involved in an anaerobic
activity, lactic acid will be accumulated in the muscles causing musculoskeletal pain and fatigue to the
exerciser (Vandewalle et al., 1987). Therefore body cannot resume activity until the lactic acid is
removed (Beaver et al., 1986). Providing a voice feedback at this time, seems to cause tainted
psychological conditions that can directly inhibit ‘motoneurons’, a neuron responsible for conveying
motor impulses (Rube & Secher, 1981). Thus the inhibition of motoneurons would degrade
performance, and produce pronounced fatigue and frustration (Rube N, Secher NH, 1980). This
psychological condition could create negative feeling towards the fitness app. Therefore, the positive
relationship between social presence in fitness app feedback and user engagement will not be
significant for anaerobic exercises. However, providing concurrent visual (e.g. electromyographic EMG) feedback during anaerobic exercise (i.e. strength training) has not shown any negative impact
on activity performance but has shown an acute enhancement in muscle strength and activation (Lucca
& Recchiuti, 1983; Ekblom & Eriksson, 2012). This could be explained by the fact that since
anaerobic activities are repetitive type of exercises performed for shorter duration (e.g. 30 seconds)
generally in a stationary position (Haskell et al., 2007) it does not involve much cognitive processes
(such as planning or decision making), thus an unobtrusive less humanized visual feedback will not
consume much cognitive resources. Therefore, the negative relationship between social presences in
fitness app feedback cognitive overload is less likely to have a significant impact for anaerobic type
activities.
In contrast, aerobic activities are performed for longer duration (i.e. at least 30 minutes). Thus, it is
highly likely that the user gets bored, feel monotony and develop negative feelings during the exercise
sessions (Van der Vlist et al., 2011). Therefore, it is essential to intrinsically motivate the user, in
order to complete the training session (Ryan et al., 1997; Schelling et al., 2009). Previous studies on
runners have shown that verbal encouragement during treadmills and outdoor running has yielded
positive outcomes (Andreacci et al., 2002) including prolonged exercise duration with improved heart
rate (Moffatt et al., 1994). Therefore, verbal encouragements can positively stimulate runners.
According to CASA (Nass et al., 1994; Nass & Moon, 2000) paradigm (as discussed above),
providing a spoken style, friendly voice feedback (such as “Way to go”, “Come on” and “Let’s go” )
can greatly enhance runner’s (or any aerobic activist’s) intrinsic motivation, involvement and
engagement. Thus we argue that, fitness apps with higher social presence can distract the user from the
boredom of aerobic exercises and enhance the user’s engagement with the fitness app. Similarly, since
aerobic activities are performed for a longer duration, providing frequent feedbacks while user is
exercising would guide the user to move forward towards the target with appropriate speed and pace
(Pryor & Kraines, 1999). Since, aerobic activity itself require various attentional resources, processing
a less humanized feedback will lead to cognitive overload in users. However, a more humanized
feedback with higher social presence can reduce user’s cognitive load and help in effectively moving
the user towards the exercise goal. Therefore, the negative relationship between social presence in
fitness app feedback and cognitive overload will be stronger for aerobic type exercisers. Hence we
hypothesize that
H5: Physical activity type (Aerobic/ Anaerobic) will moderate the relationship between social
presence in fitness app feedback and user engagement, such that the relationship will be stronger for
aerobic exercise type and insignificant for anaerobic exercise type.
H6: Physical activity type (Aerobic/ Anaerobic) will moderate the relationship between social
presence in fitness app feedback and cognitive load, such that the relationship will be stronger for
aerobic exercise types and insignificant for anaerobic exercise type.
Control variables
- workout complexity
- user demographics
- frequenting/avoiding gym
Physical activity type
- Aerobic
- Anaerobic
Social Presence in
H1
H5
fitness app feedback
-Informal Vs
3
H2
H6
-Voice Vs visual
Figure 1.
User
Engagement
H3
Cognitive Load
H4
Physical
activity
behaviour
Research Model
RESEARCH METHOD
A laboratory experiment will be conducted to test the hypotheses with 2 × 2 × 2 design (i.e. 2 groups
of feedback presentation formats (between-subject) and 2 types of physical activity (within-subject)).
The feedback presentation format will be manipulated on language style ((spoken informal vs. written
formal) and modality (voice vs. visual)). The types of physical activity will be manipulated by the
activity performed (1) aerobic (i.e. walking) (2) anaerobic (i.e. strength training). In addition to main
study variables control variable including the complexity of the workout, user demographics such as
history of using health apps, frequenting/avoiding gym will also be measured as it can affect app usage
and exercise behaviour. Measures for user engagement will be adapted from several studies(Chapman,
1997; Jones, 1998; Dellarocas & Narayan, 2006; Lehmann et al., 2012). The measures for cognitive
load will be adopted from (Lambourne, 2006). Physical activity behaviour will be measured using
mobile app with an embedded accelerometer, that can show speed and intensity of activity/body
movements (Hendelman et al., 2000). Complexity of the workout will be measured by the number of
steps involved. User demographics such as history of using health apps will be measured by the
number and duration of fitness apps used by the user. In addition, user’s habit of frequenting, avoiding
gym will also be measured. A pilot study will be conducted to refine the experimental manipulations
and assess the measurement properties. At least 180 students and staff subjects will be recruited from
academic faculties representing diverse backgrounds. Given the high level exposure to smartphones,
fitness apps, and exercise enthusiastic, these subjects can represent ideal fitness app users. According
to power analysis, 45 (180/4) subjects for each between-subject factor group can assure sufficient
statistical power of 0.8, for medium effect size(f=.25)(Cohen, 1988). Around 45 (180/4) participants
will be randomly assigned the 4 feedback presentation formats. The order by which the subjects
participate in the activity type will be randomized, such that half of the participants in each treatment
will perform aerobic first, while the other half will perform anaerobic first. To analyse the proposed
model the data stored in the prototype apps will be used. Initially MANOVA will be conducted on
dependent variable (DV) and mediators with social presence as the independent variable. Upon
significance, a series of univariate ANOVAs will be conducted on DV and mediators separately.
4
DISCUSSION
This research can have several theoretical contributions in the area of mobile health literature. First,
this study has conceptualized that CASA paradigm, cognitive load theory and social presence
literature can be combinedly used to inform the design of effective feedback mechanism for fitness
apps. Theorizing about mediating and moderating elements help us to better understand how and when
social presence in fitness app feedback can facilitate greater exercise performance. Second, the artefact
we proposed in this study has great potential to deliver a better quality fitness intervention at lower
cost, with wide reach and hence this study has important contribution to mobile health intervention
research.
The findings of this study can also offers several important implications to practice. First, it can inform
the app developers that feedback presentation format can have an important influence on user’s
engagement with the app. Second, app developers should be aware that, feedback presentation format
needs to be effectively designed based on the type of physical activity, in order to yield positive
outcomes. In particular, fitness app developers should concentrate on enhanced the social presence
features while developing the feedback mechanism of aerobic fitness apps (e.g. running apps) and
concentrate on providing more visual feedback while developing anaerobic fitness apps (e.g. strength
training apps). Third, fitness app users should choose apps that can provide appropriate feedback to
support their preferred physical activities.
In summary, this study focuses on an important design aspect (i.e. feedback mechanism) of fitness
apps that can effectively improve an individual’s physical activity performance. However, we do not
provide a comprehensive list of features and contextual factors that can promote physical activity in
fitness app users. Future studies can examine other factors that can enhance physical activity
behaviour. Besides this study only looks into two characteristics (i.e. language use and human
sounding speech) to symbolize social presence in fitness app feedback. However, future studies can
look into entire characteristics of CASA paradigm in order to gain a conclusive insights.
With obesity and overweight becoming a global health concern, we believe studies of this nature can
have important potential to improve the health and fitness of mobile app users and we hope many
scholars will follow the suit.
References
Ahtinen, A., Isomursu, M., Mukhtar, M., Mäntyjärvi, J., Häkkilä, J., & Blom, J. (2009) Designing
social features for mobile and ubiquitous wellness applications. In Proceedings of the 8th
International Conference on Mobile and Ubiquitous Multimedia. ACM, p. 12.
Andreacci, J.L., Lemura, L.M., Cohen, S.L., Urbansky, E.A., Chelland, S.A., & Duvillard, S.P. von
(2002) The effects of frequency of encouragement on performance during maximal exercise
testing. Journal of sports sciences, 20, 345–352.
Barr, P., Noble, J., & Biddle, R. (2007) Video game values: Human–computer interaction and games.
Interacting with Computers, 19, 180–195.
Beaver, W.L., Wasserman, K., & Whipp, B.J. (1986) A new method for detecting anaerobic threshold
by gas exchange. Journal of Applied Physiology, 60, 2020–2027.
Chalfonte, B.L., Fish, R.S., & Kraut, R.E. (1991) Expressive richness: a comparison of speech and text
as media for revision. In Proceedings of the SIGCHI Conference on Human Factors in Computing
Systems: Reaching Through Technology. ACM, pp. 21–26.
Chapman, P.M. (1997) Models of engagement: Intrinsically motivated interaction with multimedia
learning software.
Clark, R.C. & Mayer, R.E. (2008) E-Learning and the science of instruction: Proven guidelines for
consumers and designers of multimedia learning . San Francisco, CA: Pfeiffer.
Claussen, J., Kretschmer, T., & Mayrhofer, P. (2012) The Effects of Rewarding User Engagement-The Case of Facebook Apps. Information Systems Research, 24, 2013.
Cohen, J. (1988) Statistical Power Analysis for the Behavioral Sciences. Lawrence Erlbaum.
Connelly, K.H., Faber, A.M., Rogers, Y., Siek, K.A., & Toscos, T. (2006) Mobile applications that
empower people to monitor their personal health. e & i Elektrotechnik und Informationstechnik,
123, 124–128.
Consolvo, Everitt, Smith, & Landay (2006) Design requirements for technologies that encourage
physical activity. In Proceedings of the SIGCHI Conference on Human Factors in Computing
Systems. ACM, pp. 457–466.
Consolvo, Klasnja, McDonald, Avrahami, Froehlich, LeGrand, Libby, Mosher, & Landay (2008)
Flowers or a robot army?: encouraging awareness & activity with personal, mobile displays. In
Proceedings of the 10th International Conference on Ubiquitous Computing. ACM, pp. 54–63.
Consolvo, McDonald, Toscos, Chen, Froehlich, Harrison, Klasnja, LaMarca, LeGrand, & Libby
(2008) Activity sensing in the wild: a field trial of ubifit garden. In Proceedings of the Twentysixth Annual SIGCHI Conference on Human Factors in Computing Systems. ACM, pp. 1797–
1806.
Csikszentmihalyi, M. (1975) Play and intrinsic rewards. Journal of Humanistic Psychology,.
Csikszentmihalyi, M. & Rathunde, K. (1993) The measurement of flow in everyday life: toward a
theory of emergent motivation.
Dellarocas & Narayan (2006) A statistical measure of a population’s propensity to engage in postpurchase online word-of-mouth. Statistical Science, 21, 277–285.
Dubbert, P.M., Katell, A.D., Thompson, J.K., Raczynski, J.R., Lake, M., Smith, P.O., Webster, J.S.,
Sikora, T., & Cohen, R.E. (1984) Behavioral control of exercise in sedentary adults: Studies 1
through 6. Journal of consulting and clinical psychology, 52, 795.
Ekblom, M.M. & Eriksson, M. (2012) Concurrent EMG feedback acutely improves strength and
muscle activation. European journal of applied physiology, 112, 1899–1905.
Fujiki, Y., Kazakos, K., Puri, C., Pavlidis, I., Starren, J., & Levine, J. (2007) NEAT-o-games:
ubiquitous activity-based gaming. In CHI’07 Extended Abstracts on Human Factors in Computing
Systems. ACM, pp. 2369–2374.
Gasser, R., Brodbeck, D., Degen, M., Luthiger, J., Wyss, R., & Reichlin, S. (2006) Persuasiveness of a
mobile lifestyle coaching application using social facilitation. Persuasive Technology, 27–38.
Gefen, D. & Straub, D.W. (2004) Consumer trust in B2C e-commerce and the importance of social
presence: experiments in e-products and e-services. Omega, 32, 407–424.
Hanin, Y.L. (2007) Emotions and Athletic Performance: Individual Zones of Optimal Functioning
Model. Human Kinetics Publishers.
Haskell, W.L., Lee, I., Pate, R.R., Powell, K.E., Blair, S.N., Franklin, B.A., Macera, C.A., Heath,
G.W., Thompson, P.D., & Bauman, A. (2007) Physical activity and public health: updated
recommendation for adults from the American College of Sports Medicine and the American Heart
Association. Medicine and science in sports and exercise, 39, 1423.
Hendelman, D., Miller, K., Baggett, C., Debold, E., & Freedson, P. (2000) Validity of accelerometry
for the assessment of moderate intensity physical activity in the field. Medicine and science in
sports and exercise, 32, S442–S449.
Hess, T.J., Fuller, M., & Campbell, D.E. (2009) Designing interfaces with social presence: Using
vividness and extraversion to create social recommendation agents. Journal of the Association for
Information Systems, 10, 1.
Horvath, K. & Lombard, M. (2010) Social and Spatial Presence: An Application to Optimize HumanComputer Interaction. PsychNology Journal, 8, 85–114.
Jones, M.G. (1998) Creating engagement in computer-based learning environments. Retrieved April,
12, 2001.
Kim, K.J., Park, E., & Shyam Sundar, S. (2013) Caregiving role in human--robot interaction: A study
of the mediating effects of perceived benefit and social presence. Computers in Human Behavior,
29, 1799–1806.
Kumar, N. & Benbasat, I. (2006) Research note: the influence of recommendations and consumer
reviews on evaluations of websites. Information Systems Research, 17, 425–439.
Lambourne, K. (2006) The relationship between working memory capacity and physical activity rates
in young adults. Journal of Sports Science and Medicine, 5, 149–153.
Lee, G., Tsai, C., Griswold, W.G., Raab, F., & Patrick, K. (2006) PmEB: a mobile phone application
for monitoring caloric balance. In CHI’06 Extended Abstracts on Human Factors in Computing
Systems. ACM, pp. 1013–1018.
Lehmann, J., Lalmas, M., Yom-Tov, E., & Dupret, G. (2012) Models of user engagement. In User
Modeling, Adaptation, and Personalization. Springer, pp. 164–175.
Lin, J., Mamykina, L., Lindtner, S., Delajoux, G., & Strub, H. (2006) Fish’n’Steps: Encouraging
physical activity with an interactive computer game. UbiComp 2006: Ubiquitous Computing, 261–
278.
Lindenberger, U., Marsiske, M., & Baltes, P.B. (2000) Memorizing while walking: Increase in dualtask costs from young adulthood to old age. Psychology and aging, 15, 417–436.
Lombard, M. & Ditton, T. (1997) At the heart of it all: The concept of presence. Journal of
Computer Mediated Communication, 3, 0.
Lucca, J.A. & Recchiuti, S.J. (1983) Effect of electromyographic biofeedback on an isometric
strengthening program. Physical therapy, 63, 200–203.
Marcus, B.H., Banspach, S.W., Lefebvre, R.C., Rossi, J.S., Carleton, R.A., & Abrams, D.B. (1992)
Using the stages of change model to increase the adoption of physical activity among community
participants. American Journal of Health Promotion, 6, 424–429.
Marcus, B.H., Emmons, K.M., Simkin-Silverman, L.R., Linnan, L.A., Taylor, E.R., Bock, B.C.,
Roberts, M.B., Rossi, J.S., & Abrams, D.B. (1998) Evaluation of motivationally tailored vs.
standard self-help physical activity interventions at the workplace. American Journal of Health
Promotion, 12, 246–253.
Mattila, E., Parkka, J., Hermersdorf, M., Kaasinen, J., Vainio, J., Samposalo, K., Merilahti, J., Kolari,
J., Kulju, M., & Lappalainen, R. (2008) Mobile diary for wellness management—results on usage
and usability in two user studies. Information Technology in Biomedicine, IEEE Transactions on,
12, 501–512.
Mayer, R.E. (1984) Aids to text comprehension. Educational Psychologist, 19, 30–42.
Medbo, J.I., Mohn, A.-C., Tabata, I., Bahr, R., Vaage, O., & Sejersted, O.M. (1988) Anaerobic
capacity determined by maximal accumulated O2 deficit. Journal of Applied Physiology, 64, 50–
60.
Moffatt, R.J., Chitwood, L.F., & Biggerstaff, K.D. (1994) The influence of verbal encouragement
during assessment of maximal oxygen uptake. The journal of sports medicine and physical fitness,
34, 45.
Moore, B.C.J. & Moore, B.C. (2003) An Introduction to the Psychology of Hearing. Academic press
San Diego.
Moreno, R. & Mayer, R.E. (2000a) A learner-centered approach to multimedia explanations: Deriving
instructional design principles from cognitive theory. Interactive multimedia electronic journal of
computer-enhanced learning, 2, 12–20.
Moreno, R. & Mayer, R.E. (2000b) Engaging students in active learning: The case for personalized
multimedia messages. Journal of Educational Psychology, 92, 724–733.
Nagamatsu, L.S., Voss, M., Neider, M.B., Gaspar, J.G., Handy, T.C., Kramer, A.F., & Liu-Ambrose,
T.Y.L. (2011) Increased cognitive load leads to impaired mobility decisions in seniors at risk for
falls. Psychology and aging, 26, 253.
Napolitano, M.A., Fotheringham, M., Tate, D., Sciamanna, C., Leslie, E., Owen, N., Bauman, A., &
Marcus, B. (2003) Evaluation of an internet-based physical activity intervention: a preliminary
investigation. Annals of Behavioral Medicine, 25, 92–99.
Nass, C. & Moon, Y. (2000) Machines and mindlessness: Social responses to computers. Journal of
social issues, 56, 81–103.
Nass, C., Moon, Y., Fogg, B.J., Reeves, B., & Dryer, C. (1995) Can computer personalities be human
personalities? In Conference Companion on Human Factors in Computing Systems. ACM, pp.
228–229.
Nass, C. & Steuer, J. (1993) Voices, boxes, and sources of messages. Human Communication
Research, 19, 504–527.
Nass, C., Steuer, J., & Tauber, E.R. (1994) Computers are social actors. In Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems. ACM, pp. 72–78.
Neider, M.B., McCarley, J.S., Crowell, J.A., Kaczmarski, H., & Kramer, A.F. (2010) Pedestrians,
vehicles, and cell phones. Accident Analysis & Prevention, 42, 589–594.
O’Brien, H.L. & Toms, E.G. (2008) What is user engagement? A conceptual framework for defining
user engagement with technology. Journal of the American Society for Information Science and
Technology, 59, 938–955.
Oliver, N. & Flores-Mangas, F. (2006) MPTrain: a mobile, music and physiology-based personal
trainer. In Proceedings of the 8th Conference on Human-computer Interaction with Mobile Devices
and Services. ACM, pp. 21–28.
Paas, F., Tuovinen, J.E., Tabbers, H., & Van Gerven, P.W.M. (2003) Cognitive load measurement as a
means to advance cognitive load theory. Educational psychologist, 38, 63–71.
Pate, R.R., Pratt, M., Blair, S.N., Haskell, W.L., Macera, C.A., Bouchard, C., Buchner, D., Ettinger,
W., Heath, G.W., & King, A.C. (1995) Physical activity and public health. JAMA: the journal of
the American Medical Association, 273, 402–407.
Pryor, E. & Kraines, M.G. (1999) Keep Moving!: Fitness Through Aerobics and Step. Mayfield
(Mountain View, Calif.).
Rafaeli, S. (1990) Interacting with media: Para-social interaction and real interaction. Mediation,
information, and communication: Information and behavior, 3, 125–181.
Reeves, B. & Nass, C. (1996) How People Treat Computers, Television, and New Media Like Real
People and Places. CSLI Publications and Cambridge university press.
Roda, C. (2011) Human Attention in Digital Environments. Cambridge University Press.
Rube N, Secher NH, L.F. (1980) Training and the paradoxical influence of verbal encouragement on
muscle fatigue. Acta Physiol Scand, 108.
Rube, N. & Secher, N.H. (1981) Paradoxical influence of encouragement on muscle fatigue. European
Journal of Applied Physiology and Occupational Physiology, 46, 1–7.
Ryan, Frederick, Lepes, Rubio, & Sheldon (1997) Intrinsic motivation and exercise adherence.
International Journal of Sport Psychology, 28, 335–354.
Schelling, S., Munsch, S., Meyer, A.H., Newark, P., Biedert, E., & Margraf, J. (2009) Increasing the
motivation for physical activity in obese patients. International Journal of Eating Disorders, 42,
130–138.
Short, J., Williams, E., & Christie, B. (1976) The social psychology of telecommunications.
Sproull, L., Subramani, M., Kiesler, S., Walker, J.H., & Waters, K. (1996) When the interface is a
face. Human-Computer Interaction, 11, 97–124.
Sweller, J. (1994) Cognitive load theory, learning difficulty, and instructional design. Learning and
instruction, 4, 295–312.
Sweller, J., Van Merrienboer, J.J.G., & Paas, F.G.W.C. (1998) Cognitive architecture and instructional
design. Educational psychology review, 10, 251–296.
Tate, D.F., Wing, R.R., & Winett, R.A. (2001) Using Internet technology to deliver a behavioral
weight loss program. JAMA: the journal of the American Medical Association, 285, 1172–1177.
Tsai, C.C., Lee, G., Raab, F., Norman, G.J., Sohn, T., Griswold, W.G., & Patrick, K. (2007) Usability
and feasibility of PmEB: a mobile phone application for monitoring real time caloric balance.
Mobile networks and applications, 12, 173–184.
Tu, C.-H. & McIsaac, M. (2002) The relationship of social presence and interaction in online classes.
The American journal of distance education, 16, 131–150.
Tung, F.-W. & Deng, Y.-S. (2006) Designing social presence in e-learning environments: Testing the
effect of interactivity on children. Interactive learning environments, 14, 251–264.
Turgut, Y. & Irgin, P. (2009) Young learners’ language learning via computer games. Procedia-Social
and Behavioral Sciences, 1, 760–764.
Vallerand, R.J. & Blanchard, C.M. (2000) The study of emotion in sport and exercise. Emotions in
sport, 3–37.
Van der Vlist, B., Bartneck, C., & Mäueler, S. (2011) moBeat: Using interactive music to guide and
motivate users during aerobic exercising. Applied Psychophysiology and Biofeedback, 36, 135–
145.
Vandewalle, H., Péerès, G., & Monod, H. (1987) Standard anaerobic exercise tests. Sports Medicine,
4, 268–289.
Vealey, R.S. (2005) Coaching for the Inner Edge. Fitness Information Technology.
Verghese, J., Buschke, H., Viola, L., Katz, M., Hall, C., Kuslansky, G., & Lipton, R. (2002) Validity
of divided attention tasks in predicting falls in older individuals: a preliminary study. Journal of the
American Geriatrics Society, 50, 1572–1576.
Walther, J.B. (1992) Interpersonal Effects in Computer-Mediated Interaction A Relational Perspective.
Communication research, 19, 52–90.
Winograd, T.A. & Flores, C.F. (1986) Understanding Computers and Cognition: A New Foundation
for Design. Ablex Pub.
Witmer, B.G. & Singer, M.J. (1998) Measuring presence in virtual environments: A presence
questionnaire. Presence, 7, 225–240.
Yoganathan, D. & Kajanan, S. (2013) Persuasive Technology for Smartphone Fitness Apps.