Blog

Workshop: Eye-Tracking

Eye-tracking as an experimental tool

This workshop will introduce participants to eye-tracking as an experimental method. Tracking eye movements—where and how long a people look at stimuli, or surroundings—is gaining momentum as a way to trace cognition through attention patterns.

Part 1: How eye-tracking is conducted

Location: Visual Sciences Lab, Room 3.739, Centre for Brain Science

The workshop will begin with demonstrations of eye-tracking, and how it is used to tackle a variety of research questions. There will be the opportunity to experience an eye-tracking experiment from both experimenter and participant perspectives.

Presenters:
Kate McCulloch: An introduction and overview to eye-tracking
Luke Holmes: Pupil dilation to explicit sexual stimuli
Dawn Liu: Fixations on food labels
Sarah Edwards: The use of mobile eye-tracking in attention to corridor posters

— Tea/Coffee Break —

Part 2: Eye-tracking data across disciplines

Location: Computer lab 2.708, Psychology Building (Square 1)

In the second half, the workshop will cover applications of eye-tracking in disciplines such as accounting, art, biology, economics, linguistics, and more. We will show what data comes from an eye-tracking experiment and how that can be analysed quantitatively and descriptively.

The workshop will end with free-roaming time where participants can explore eye-tracking experiments, work with sample eye-tracking data, create eye-tracking visuals, and get advice on how to integrate the methodology into their own research.

Presenters:
Kate McCulloch: Building a simple eye-tracking experiment
Dawn Liu: Uses of eye-tracking: Cross-disciplinary applications
Dawn Liu: Working with eye-tracking data

Materials for the workshop are here: https://github.com/dliuxi/ECEM

Advertisements

Paper Presentation: John of Gaunt

June 21, 16.20-16.50: Jessica Fure, University of Essex

John of Gaunt: The Fictions of History/A History of Fictions
An examination of the use of John of Gaunt in Literature and the Arts

My research looks at how an impression of John of Gaunt, a real person, is shaped through a progression of fictional portrayals in both literature and art, with a special focus on major works from 1730-1950. Although Gaunt appears consistently in various works during his life and after his death, from the early Georgian era onwards these appearances become driven towards the public in the manner of all commercial art, fostering a consumption of image and identity, while expanding the popular culture’s consciousness of the historical person to suit the artists’ aims.

Beginning with a short background in the Tudor era, with Shakespeare’s Richard II and the earliest known portraits of Gaunt, I proceed onward to the Georgian era, where consuming culture became a political act available to the general public and where the idea of a British identity could be, in a sense, purchased and displayed. From there, Ford Maddox Brown utilizes Gaunt to advance his idea of Victorian moral obligations, and the mid-twentieth century American author Anya Seton uses the Duke to speak to an audience looking for women’s experience in postwar fictions.

These works engage fully with the available popular histories of the time, to question and embellish scholarly history, and to employ previous “common knowledge” regarding Gaunt – including mining previous uses in the arts – each creating, in the end, a cumulative portrait that accomplishes what T. S. Eliot claimed was an artist’s ultimate goal: to find the unspoken idea or unexplained idea and to articulate that into humanity’s ongoing conversation, to add to what we know in a manner that speaks to having listened.

The interdisciplinary aspect of the work looks at the intertwining methods and reasons for creating a character for public consumption, and the results of that consumption on later works. My analysis methods cross techniques from literature, history, art history, and film studies.  There is a strong element of close reading to emphasise how the particular texts or paintings articulate intended goals and/or highlight the ways the works are anchored in their time.

The additional aim of my project is to show the value in an interdisciplinary approach to any subject – to view the artefacts as objects in a museum collection that can be utilized in several different contexts, and to provide a case study in how to come to a subject from several angles and highlight myriad avenues outward.

My true passion is for interdisciplinary education, encouraging students in different disciplines to use the same subject, coming together to express differing viewpoints, using their particular skillsets and interests and sharing those, leading to new, exciting ways of seeing that subject. Crossing these boundaries, whether it leads to continued collaboration, or simply enhances one’s own studies, leads to a richer experience and the potential for more far-reaching research.  To that end, my thesis is to a degree, a case study for this multi-purposing of a subject, done on a single-student scale.

Paper Presentation: Pupil Dilation as a Measure of Sexual Arousal

June 21, 15.50-16.20: Luke Holmes, University of Essex

Pupil Dilation as a Measure of Sexual Arousal

Background: Pupil dilation to explicit sexual stimuli (footage of naked and aroused men or women) appears to be a valid indicator of sexual arousal, because responses reflect sex differences and sexual orientation differences previously described with the (more invasive) measure of genital arousal. However, viewing explicit sexual stimuli can be invasive in itself and limit researchers in recruiting a diverse range of participants. If previous patterns of results were replicated with non-explicit sexual stimuli (footage of dressed men and women), then pupil dilation could be indicative of automatic sexual response in fully non-invasive designs.

Hypotheses: Based on previous literature on both genital arousal and pupil dilation we expect the following patterns, if pupil dilation to non-explicit sexual stimuli is a valid indicator of sexual response: 1. Pupil dilation to explicit sexual stimuli depicting males or females will relate to pupil dilation to corresponding non-explicit sexual stimuli. 2. For both explicit and non-explicit sexual stimuli, the correspondence of pupil dilation to male or female sexual stimuli with self-reported sexual orientation will be stronger in men than women. 3. For both explicit and non-explicit sexual stimuli, the concordance of pupil dilation to sexual stimuli with time spent viewing these stimuli, and self-reported sexual attraction to these stimuli, will be stronger in men than women. 4. For both types of stimuli, the relationships of pupil dilation with self-reported sexual orientation, viewing time, and self-reported sexual attraction to stimuli will be significant, although effects will be stronger for explicit stimuli than non-explicit stimuli.

Method: Participants were 165 men and 160 women of different sexual orientations whose pupil dilation and viewing time were measured with an SR Research infrared eye tracker. Sexually explicit stimuli were 24 videos depicting either a male or a female model masturbating; non-explicit stimuli were 24 videos of a clothed male or female discussing the weather. Videos were 30 seconds long. For the assessment of pupil dilation, participants watched videos before answering three questions regarding their sexual attraction to the depicted person. For the assessment of viewing time, and for each stimulus type (explicit or non-explicit) participants viewed a male stimulus next to a female stimulus, and the time viewing one over the other was recorded. Correlations and linear regression analyses were conducted to test hypotheses.

Results: 1. Dilation patterns to non-explicit stimuli modestly but significantly resembled those to explicit stimuli depicting the same sex or other sex. 2. With explicit stimuli, sexual orientation differences in pupil dilation were significantly larger in men than in women. However, with non-explicit stimuli, this pattern was not confirmed. 3. For both types of stimuli, correlations between pupil dilation, viewing time, and self-reported sexual attraction were significantly stronger in men than women. 4. In general, correlations between pupil dilation and other responses to stimuli were stronger for explicit stimuli than non-explicit stimuli, although they were significant for both types of stimuli.

Conclusions: Pupil dilation patterns to explicit and non-explicit sexual stimuli were modestly related, and most (but not all) predicted sex differences and sexual orientation differences were detected with both types of stimuli. Overall, the present data support the proposal that the assessment of pupil dilation can be combined with non-explicit stimuli for non-invasive studies on sexual attraction and arousal. In several traditional cultures it is impossible to assess genital arousal or to show sexually explicit footage. Thus, the advent of mobile eye trackers, in combination with non-explicit sexual stimuli, shall allow assessing sexual responses in these cultures in a non-offensive manner.

Funding Sources: American Institute of Bisexuality, US Department of Agriculture

Paper Presentation: Dominated Strategies in Co-ordination Games

June 21, 12.20-12.50: Atiyeh Yeganloo, The University of Manchester

A Useful Bias: the Role of Dominated Strategies in Coordination Game

In games with multiple Nash equilibria, coordination failure can lead to off-equilibria and (Pareto) dominated outcomes. I use the dominance effect, a commonly observed bias in individual choices, to facilitate coordination among players in games.

Dominance effect predicts that the addition of a dominated alternative increases the psychological attractiveness of the dominating alternative.
Whereas, dominated alternatives are irrelevant under the rationality assumption.

A dominated strategy in games can affect both players’ strategy choices and facilitate coordination toward one of Nash equilibria of the games.
Results from an experiment on the 2×2 Battle of Sexes games, with a dominated strategy for row players, is consistent with this prediction.

Observed choices of row players differ depending on the presence of the dominated strategy. Column players anticipate row players’ choices when the dominated strategy presents. Column players coordinate on their strategy choices to reach to the relevant equilibrium.

Further, I study if the dominance effect carries over from risky choices to games. That is, players biased by a dominated lottery in risky choices are more likely to show the bias in games. This bias is shown to be present in risky choices. There is no correlation at individual level between games and lotteries regarding the dominated alternative and its effect.

Paper Presentation: Peer Effects and Social Closeness

June 21, 11.50-12.20: Georg Sator, University of Vienna

Peer Effects and Social Closeness

We introduce social closeness to the economic literature on peer effects by studying how their strength and magnitude depends on how closely a decision-maker identifies with its peer.

There is a large psychological literature on what characterizes social closeness. We work with the definition of unilateral subjectively perceived closeness to another person. Most importantly this is sharply distinct from social proximity, which aims at how many, what kind of or how important personal characteristics someone may share with someone else.

We speak of peer effects if the mere presence of other people or knowledge of their behaviour, changes a decision-maker’s preferences and hence her choices. We carefully demarcate peer effects conceptually and empirically from influences stemming from other sources of interaction such as strategic considerations, information spillovers or signalling.

A thorough understanding of peer effects in general, their magnitude and the circumstances under which they can be expected to arise is important because of the economic significance of the settings in which they have been shown to play a role. Peer effects have been shown to be a crucial ingredient in determining investments in education, effort provision at the workplace, charitable giving or the sustainability of cooperation when institutions are weak. 

Despite the fact that there is plenty of evidence for the relevance of peers for people’s choices, little is known about the conditions under which we can expect them to occur. Since peer effects rely on reputational concerns and imitation and both of these concepts are likely to be affected by who the other person is and how she relates to oneself, we think that social closeness between peers is a likely candidate for their occurrence. So far peer effects have been studied in (typically mutually anonymous) lab settings, where the decision-maker had no information about her peers and how she compares to herself or in field settings with little control over the social closeness between peers.

We use an experimental design that allows us to systematically vary social closeness between peers and hence study its effect on choices. On top of that we isolate the two channels though which peer effects can operate: first, the effect on an observed agent, who may make a different choice due to the fact that the peer can see what she is doing, or second, the effect on the observing agent, who, prior making her own choice, receives information about what her peer did in an identical situation.

A key feature of our design constitutes the unusual experimental procedure. We recruited subjects pairwise with a friend and moreover, we introduce them to a stranger prior to the start of the actual experiment. In the course of the experiment, interaction between peers then takes place between either a pair of friends, strangers, anonymous subjects, or in the control, not at all.

Since we are interested in cooperation we use a standard linear public good game.

We find that peer effects are indeed affected by social closeness.

Observed subjects choose more prosocially than subjects who choose in private. Interestingly, when differentiating with respect to social closeness, we see that this general result is exclusively driven by subjects who are observed by people they do not or hardly know. If observations are made between agents who know is other very well from outside the lab the effect completely vanishes.

With respect to those subjects who observe a peer’s decision prior to making their own decision we find that their choices are more similar to the choice they have observed that two subjects’ choices are when there was no observation between them. Differentiating with respect to social closeness reveals that this effect is driven by those subjects who observed socially close people.

In the light of these results we argue that the occurrence of peer effects depends on the social closeness between the peers in the following way: The drivers of peer effects seem to be reputational concerns vis-à-vis strangers and imitation of friends. We conduct multiple robustness checks to rule out alternative explanations. Most notably we control for similarity of socially close people, such as friends, in unobserved characteristics, which is a notorious difficulty in research on peer effects.

Paper Presentation: Field Experiments in Restaurants

21 June, 11.20-11.50: Merav Malcman, Ben-Gurion University

Field Experiments in Restaurants: The Effects of Music

Scientific Background and Working Hypotheses
Music has a potential influence on customers’ behavior and decision-making. Background music affects behavior in a variety of ways, and previous research has tried to understand how people react in everyday commercial environments (e.g. restaurants, markets, wine store, etc.), and which emotions are involved in each situation (Milliman, 1986; Wilson, 2003). Moreover, the effect of music depends on the nature of the task and the type of music playing in the background.

Many terms are used to describe music; three basic factors are time, pitch and texture (Bruner, 1990). Time includes, inter alia, rhythm (the pattern of accents given to beats or notes in a song) and tempo (the speed or rate at which a rhythm progresses). Pitch includes the melody and mode of the music. Melody is the succession of notes occurring over time throughout a song. Changes in melody can be either ascending (rising in pitch) or descending (dropping in pitch). Mode refers to the series of notes, arranged in a scale of ascending pitch that provides the tonal substance of a song. The best-known modes are the major and the minor modes. The texture of music refers to its timbre and volume. Timbre describes the distinctive tone that makes one instrument sound different from another even if they are both playing the same melody at the same pitch. Volume also contributes to the texture of music; for example, by making one note louder than others around it.

The literature suggests that tempo is one of the most important determinants of human response to music. Milliman (1986) showed the effect of music tempo on the behavior of restaurant patrons. When slow music was playing, customers spent significantly more time in the restaurant and more money on alcoholic beverages. Caldwell & Hibbert (1999) also found that with a slow tempo, patrons spent significantly more time on dining; although there was a clear trend towards higher spending in their slow tempo group, the amount spent on food, not only beverages, was also significant.

Tempo is strongly correlated with arousal. Fast music has been shown to raise listeners’ self-reported arousal levels (Knoferle, Spangenberg, Herrmann, & Landwehr, 2012). Tempo also effects time perception (Caldwell & Hibbert, 1999; Knoferle et al., 2012; Milliman, 1986) so diners in the slow tempo condition may underestimate time spent in the restaurant.

This study use the musical tempo as an independent variable to report its influence on time spent, bill amount and the tip size and accordingly the hypotheses are:

H1: When the tempo (speed) of the music is slow (fast), patrons spend more (less) time dining.

H2: When the tempo (speed) of the music is slow (fast), patrons spend more (less) time dining and their pleasure is positively aroused, hence the bill amount will be higher (lower).

H3: When the tempo (speed) of the music is slow (fast), patrons spend more (less) time dining and their pleasure is positively aroused, hence the tip size will be higher (lower).

In this study, the experiment is conducted without prior knowledge of patrons’ music preferences. The research question in this study is what are the effects of music tempo on tipping?

Experimental Design & Methods
Manipulations
The two tempo conditions were created based on the criteria used by Milliman (1986) and reused by Caldwell & Hibbert (1999) in their research: music with 94 or more beats per minute was used for the fast tempo condition, while music with 72 or fewer beats per minute was used for the slow tempo condition. The manipulation will be to play music with a fast tempo, denoted FT , or with a slow tempo, denoted ST . The volume of the music will remain consistent throughout all manipulations.
In order to acquire better control on the music variable, the tempo will be randomly assigned on different days of the week, and reversed the following week. In the control group, the music will be the regular music played at the restaurant. The playlist will be a combination of female and male vocalists, both well-known popular artists and lesser known, less popular ones. The music (songs) for all manipulations will be chosen by a company that specializes in playlists for background music in restaurants, and with whom the restaurant has an ongoing contract. The Musical genre was Italian music and Greece music, that are the regular genre of the background music in the restaurant.

Data Collection and Experimental Procedures
The data will be collected from a restaurant in Tel Aviv, over a period of several months, in different seasons, and on different days of the week. All data will be from diners served from 17:00 to 00:00, in parties of 1-2 adults without children or with up to 3 children. The common practice in this restaurant is that all tips are pooled and shared between the waiters at the end of each session, prorated by the number of hours worked. This practice will be kept for the tips collected during the experiment.

At the beginning of each session, each day, these instructions will be relayed to the relevant servers: I’m collecting statistical data about the restaurant. I will request specific data on activity at tables with 1-2 adult patrons per table or 1-2 adult patrons with up to and including 3 children. Once the patrons have left the restaurant, you will be required to provide me with information about the bill and your tip. Your cooperation will be greatly appreciated.

To ensure that the sampling of the target population of the research is random, a timetable of predetermined manipulations will be set, and different manipulations will alternate by days of the week. For example, if the study begins with manipulation FT on a Sunday, the following Sunday will continue with manipulation ST, and so on for all dates.
The main data variables that collected are: Treatments (Regular = no manipulation = the regular music on the restaurant, FT = fast tempo music, ST = slow tempo music), Bill amount, Tip amount, Time of entrance, Time of paying the bill and more variables. The complete data variables will be entered into the database, as explained in Appendix.

Bibliography
Bruner, G. C. (1990). Music, mood, and marketing. the Journal of marketing, 94-104.
Caldwell, C., & Hibbert, S. A. (1999). Play that one again: The effect of music tempo on consumer behaviour in a restaurant. European Advances in Consumer Research, 4, 58-62.
Husain, G., Thompson, W. F., & Schellenberg, E. G. (2002). Effects of musical tempo and mode on arousal, mood, and spatial abilities. Music Perception: An Interdisciplinary Journal, 20(2), 151-171.
Knoferle, K. M., Spangenberg, E. R., Herrmann, A., & Landwehr, J. R. (2012). It is all in the mix: The interactive effect of music tempo and mode on in-store sales. Marketing Letters, 23(1), 325-337.
Milliman, R. E. (1986). The influence of background music on the behavior of restaurant patrons. Journal of consumer research, 286-289.
Wilson, S. (2003). The effect of music on perceived atmosphere and purchase intentions in a restaurant. Psychology of music, 31(1), 93-112.

Paper Presentation: Age differences in risk-taking behaviour

21 June, 10.50-11.20: Kelly Wolfe, University of Essex

Age differences in risk-taking behaviour: a matter of risk-attitude or decision-quality?

Introduction: Previous research examining age differences in decision-making under risk has yielded mixed findings. Some studies have reported greater risk-taking in older age compared to younger adults, other studies have reported less risk-taking in older age, and other studies have reported no age differences. What factors underlie these mixed findings regarding adult age differences in risk-taking tendencies? We propose that the mixed findings result at least partly from the multitude of different methods and materials used to measure risk-taking behaviour. Specifically, we propose that some tasks used to assess risk-taking tendencies may be more cognitively demanding or require more proficient numeracy skills than other tasks, and as such, load more heavily on cognitive abilities and numeracy skills that decline in older age. Thus, age differences in decision-making on risk-taking tasks may result from both or either (a) age differences in decision-making quality due to age-related decline in cognitive abilities and numeracy and (b) age differences in attitudes towards risk-taking. The aim of our research is to disentangle the contribution of decision-making quality and risk-taking attitude to age differences in decision-making under risk.

Method: Participants evaluated 10 monetary gambles and estimated what they believed to be the expected value of each gamble. They were asked to do so by mimicking the probabilities of the outcomes associated with each gamble in a box with 20 compartments. Each gamble had three possible outcomes; a probability to win, a probability to lose, and a probability to neither win nor lose. For each gamble, participants were asked to identify the frequency of win, loss, and neither win nor lose outcomes imagining that the gamble were played 20 times. This method assessed participants’ understanding of the probabilities associated with the three possible gamble outcomes (i.e., win, lose, neither win nor lose).  Following completion of this task, participants received feedback about their judged expected value of each gamble based on the frequency that they determined for each possible outcome. Participants were then asked to decide whether they wished to play each of the gambles. Following all decisions, participants received as monetary payment any winnings they had accrued across the 10 gambles. Cognitive ability was assessed using measures of processing speed and backward digit span. Numeracy was assessed with an 11-item numeracy scale. Positive and negative affect and self-reported risk-taking were also assessed.

Results: Correct judgements of the probability to win and lose were associated with numerical ability, irrespective of age. Incorrect judgments of the probability to win and lose were associated with age, as older age predicted a larger distance between participant’s estimation and the actual probabilities to win and lose. A higher likelihood of accepting to play a gamble was associated with incorrectly judging the probability of winning as higher than the actual probability to win or incorrectly judging the probability of losing as lower than the actual probability to lose. Also, higher numerical ability was also associated with a lower likelihood of accepting gambles. Age was not associated with likelihood of accepting to play gambles.

Conclusion: Our findings suggest that numerical abilities play a large role in assessing the value of a monetary gamble, irrespective of age, nor did younger and older adults differ in the likelihood of accepting these gambles. However, when probabilities to win and lose were incorrectly estimated, older age was related to a larger distance between the estimated and the actual probability to win and lose.