Eyes keep watch over you! Competition enhances joint attention in females
Francesca Ciardo1, Paola Ricciardelli2,3, Luisa Lugli4, Sandro Rubichi1, Cristina Iani1
(1) Department of Communication and Economics, University of Modena and Reggio Emilia Viale Allegri, 9, 42121 Reggio Emilia, Italy
(2) Department of Psychology, University of Milano-Bicocca, Edificio U6, Piazza dell’Ateneo Nuovo 1, 20126 Milan, Italy
(3) Milan Centre for Neuroscience, Piazza dell’Ateneo Nuovo 1, 20126, Milan, Italy (4) Department of Philosophy and Communication, University of Bologna, University of Bologna,
Via A. Gardino 23, 40122 Bologna, Italy
Word count (main text): 6287 References: 51
Address for Correspondence: Francesca Ciardo or Cristina Iani
Department of Communication and Economics, University of Modena and Reggio Emilia Viale Allegri, 9, 42121 Reggio Emilia, Italy
ABSTRACT
The present study investigated if the gaze-cuing effect (i.e., the tendency for observers to respond faster to targets in locations that were cued by others’ gaze direction than to not-cued targets) is modulated by the type of relationship (i.e., cooperative or competitive) established during a previous interaction with a cuing face. In two experiments, participants played a series of single-shot games of a modified version of the two-choice Prisoner's Dilemma against eight simulated contenders. They were shown a fictive feedback indicating if the opponents chose to cooperate or compete with them. Opponents’ faces were then used as stimuli in a standard gaze-cuing task. In Experiment 1 females classified as average in competitiveness were tested, while in Experiment 2 females classified as high and low in competitiveness were tested. We found that only in females classified as low and average in competitiveness the gaze-cuing effect for competitive contenders was greater than for cooperative contenders. These findings suggest that competitive opponents represent a relevant source of information within the social environment and female observers with low and average levels of competition cannot prevent from keeping their eyes over them.
1. INTRODUCTION
Orienting attention in the same direction where other individuals are looking at, known as joint attention, is a foundational ability guiding social interactions and communication (Langton, Watt, & Bruce, 2000). The gaze-mediated orienting of attention allows the observer to infer other people’s mental states (i.e., the focus of their attention; e.g. Baron-Cohen, 1995), and to understand and predict their future actions (e.g., Pierno et al., 2006; Innocenti, De Stefani, Bernardi, Campione, & Gentilucci, 2012).
Even though there are several indications of the automatic and reflexive nature of the mecha-nisms underlying the gaze-mediated orienting of attention (e.g., Galfano et al., 2012; Hayward & Ristic, 2013; but see Frischen, Bayliss, & Tipper, 2007 for a review), recent studies suggested that joint attention orienting may not be purely bottom-up driven, but it is rather influenced by top-down processes that interpret the averted gaze based on its relevance for the task (e.g., Ricciardelli, Carcagno, Vallar, & Bricolo, 2013), and in comparison to other stimuli in the environment (e.g., Green, Mooshagian, Kaplan, Zaidel, & Iacoboni, 2009; Ristic & Kingstone, 2005).
Furthermore, there are recent reports that joint attention is influenced by implicit and explicit social information associated to the seen person, such as age (Ciardo, Marino, Actis-Grosso, Ros-setti, & Ricciardelli, 2014), facial emotional expression (e.g., Bonifacci, Ricciardelli, Lugli & Pelli-cano, 2008), gender (Olhsen, van Zoest, & van Zugt, 2013), race (Pavan, Dalmaso, Galfano, & Castelli, 2011), social status (e.g., Dalmaso, Pavan, Castelli, & Galfano, 2012) and political affilia-tion (Liuzza et al., 2011). For instance, Liuzza and colleagues(2011), using an oculomotor task, re-ported that political affiliation enhanced joint attention for in-group voters while it inhibited it for out-group voters. Similar results have been reported by Pavan et al. (2011) who investigated the im-pact of racial group membership on covert orienting of joint attention and reported that White par-ticipants selectively shifted their attention in response to the averted gaze of own-race individuals only. Therefore, person categorization seems to regulate joint attention.
The way we perceive and categorize others may depend on the type of relationship we have with them. Group membership acts as a strong categorization cue. There are indications that the need to cooperate or to compete leads to in-group/out-group differentiations, with individuals per-ceiving themselves as part of the same social group when they need to cooperate with each other. On the contrary, when the others represent an obstacle toward the attainment of a goal, they are more likely perceived as out-group members (e.g., Rabbie & Horwitz, 1969). In general, competi-tive interactions have been shown to increase perceived intergroup and interpersonal differences (e.g., Toma, Yzerbyt, & Corneille, 2010) and to disrupt the emergence of shared task representa-tions in joint action (e.g., Iani, Anelli, Nicoletti, Arcuri, & Rubichi, 2010). Interestingly, the effects of competition may be long lasting and transfer from one task to a subsequent one (e.g., Iani, Anelli, Nicoletti, & Rubichi, 2014; Sassenberg, Moskowitz, Jacoby, & Hansen, 2007).
To note, competitors are a special kind of out-group members as the achievement of their goals often coincides with the impossibility to reach ours. For these reasons, in competitive situa-tions it appears fundamental for the observer to monitor their progress toward their goals (Poortvliet & Darnon, 2010). Evidence indicates that social interactions characterized by cooperation differ-ently affect individual cognitive processes, such as attention, face recognition and memory, with faces of non-cooperative individuals attracting more automatic attention (e.g., Vanneste, Verplaet-seb, Van Hiela, & Braeckmanc, 2007) and being memorized more accurately (e.g., Chiappe et al., 2004; Mealy, Daood, & Krage, 1996; Oda, 1997; Yamagishi, Tanida, Mashima, Shimoma, & Kanazawa, 2003) than those of cooperative individuals.
Given the above considerations, the present study aimed to better clarify how social informa-tion, in particular person categorizainforma-tion, influences joint attention. Since previous studies indicated that females show stronger gaze cuing effects (Bayliss, Di Pellegrino, & Tipper, 2005) and higher sensibility to social cues (e.g., Deaner, Shepherd, & Platt, 2007; see Geary, 2010 for a review) than males, to avoid additional sources of variability in the data, only female participants were included in the present study (see Bayliss, Schuch, & Tipper, 2010). Specifically, we run two experiments to
investigate whether the gaze-cuing effect is modulated by the cooperative and competitive behavior associated to a cuing face during a prior interaction. In Experiment 1, we included only individuals classified as average in competition, as measured through the Competitiveness Index Questionnaire developed by Smither and Houston (1992). In Experiment 2, participants rated as high and low in competitiveness were included.
In both experiments, we employed a standard gaze-cuing paradigm (Driver et al., 1999) and manipulated the relationship between the participant and the cuing faces via a preliminary learning phase in which participants associated faces of simulated opponents with a cooperative or competi-tive behavior. In this phase, participants were required to play a series of single-shot games of a modified version of the two-choice Prisoner's Dilemma against eight simulated contenders. After participants had indicated their response at the social dilemma game, they were shown a fictive feedback indicating if the opponents chose to cooperate or compete with them. Opponents’ faces were then used as stimuli in the subsequent standard gaze-cuing task.
2. EXPERIMENT 1
The present experiment aimed at investigating whether the gaze-cuing effect (i.e., the ten-dency for observers to respond faster to targets in locations that were cued by others’ gaze direction than to not-cued targets) is modulated by the type of relationship (i.e., cooperative or competitive) established during a previous interaction with a cuing face. Since individuals’ competitiveness lev-els may affect how situations and others are perceived (e.g., Houston, Kinnie, Lupo, Terry, & Ho, 2000), in this experiment we included only individuals classified as average in competition. We hypothesized that cooperative and competitive interactions between the observer and the seen face differently affect joint attention. Specifically, if a cooperative behavior associated with the seen face influences joint attention by increasing group membership, as suggested by the results of previous studies on joint attention (e.g., Liuzza et al., 2011; Pavan et al., 2011), then the gaze-cuing effect should be magnified for cooperative individuals which tend to be classified as in-group
members. On the contrary, the gaze-cuing effect should not be evident for faces associated with a competitive behavior, since they are classified as out-group members (Rabbie & Horwitz, 1969). Alternatively, if competition affects joint attention by increasing the salience of some averted gazes more than others, then the gaze-cuing effect should be magnified for faces associated with
competitive outcomes, since they represent a privileged source of information in the environment. In both cases, we expected these modulations to affect longer response times since information about the prior interaction with the cuing face has to be retrieved from memory. To this aim, response times were analyzed by means of the bin distributional analysis (e.g., Ratcliff, 1979). The distributional analysis of reaction times (RTs) consists in rank ordering individual RTs as a function of the experimental conditions and in dividing them into a number of intervals or bins so that it can be seen whether a variable influences different portions of the RT distribution (e.g., De Jong, Liang, & Lauber, 1994; see Balota, & Yap, 2011 for a review). In the present study we used this technique to assess the time course of the social manipulation (i.e., the type of interaction) on the gaze-cuing effect. Specifically, if the modulation of the gaze-cuing effect is due to the retrieval from memory of the social information, then the difference between competitive and cooperative relations on the gaze-cuing effect should emerge for slower responses (i.e., the longer intervals of the distribution).
2.1 METHODS
2.1.1 Participants
Twenty-four female undergraduate students (1 left-handed; M = 21.08 years, SD = 2.21 years) from the University of Modena and Reggio Emilia received course credit to participate to the study. All had normal or corrected-to normal vision and were naïve as to the purpose of the experiment. The study was conducted in accordance with the ethical standards laid down in the Declaration of Helsinki, and fulfilled the ethical standard procedure recommended by the Italian Association of Psychology (AIP). Written consent was obtained for all of them and they were debriefed about the study at the end of the experiment.
They were selected from a larger sample (n=95) based on their scores at the Competitiveness Index (CI) questionnaire (Smither & Houston, 1992; Houston et al., 2000). To be included in the study, they had to score in the average competitiveness range (4-12). Mean scores for the partici-pants included in the study were 8.17 (SD = 2.55).
Apparatus and Stimuli
Each participant was tested in a single session, lasting about 45 minutes. All tasks were performed on a computer. Participants sat approximately 55 cm from a 17’’ CRT screen driven by a 700 MHz processor computer. Stimulus presentation and data collection were controlled by a PC running E-Prime version 2.0 software system (Psychology Software Tools, Inc).
Simulation of social interactions was achieved using a modified version of the two-choice Prisoner’s Dilemma game (i.e., Poundstone, 1992), that provides one of the most widely used way to induce cooperative and competitive outcomes (Chan & Ybarra, 2002; Vonk, 1998). The dilemma (translated here from Italian) was presented as follows: "You and another student have copied the
final test of the Psychology class. The Professor has noticed that your tests are identical and he convenes separately both of you into his office with no means of speaking to or exchanging messages between the two of you. The Professor tells you that: a) If both of you confess that you have copied the test, then you will have to wait for 6 sessions before taking the test again; b) If neither of you confesses, then both of you will have to wait for 1 session before taking the test again; c) If only one of you confesses, he/she will pass the exam without taking the test again, while the other has to wait for 7 sessions before taking the test again”.
Stimuli used in the social dilemma, in the memory test and in the gaze-cuing task were grayscale photographs (7.98° × 15.76°) depicting 8 young adults (4 females and 4 males) bearing a neutral expression. All photographs were taken from the Productive Aging Lab Face Database (Minear & Park, 2004). In the gaze-cuing task, for each photograph, gaze direction was manipulated using Adobe Photoshop so that gaze could be averted 0.65° to the left or to the right.
2.1.2 Procedure
At the beginning of the experimental session, participants were told that they would take part in three different and independent tasks concerning social judgment, memory and attention. This general introduction allowed us to administer the social dilemma and social interaction rating procedures before the gaze-cuing task. Participants were first required to play a series of single-shot games of the modified version of the two-choice Prisoner's Dilemma against eight simulated contenders. They were presented with photographs depicting the 8 young adults (i.e., opponents) bearing a neutral expression and a straight gaze. For each photograph, participants were required to imagine of being in the situation described by the social scenario and to indicate if they wanted to cooperate or compete with the opponent. Participants were provided with the cover story that the day before the opponents performed the same task seeing the participant student-id photograph. For each game, participants were shown the text of the social dilemma, their possible choices and the associated outcomes, the photograph of their opponent and a fictive name associated with the face. Half of the participants pressed the B-key on a standard keyboard if they decided to confess and the H-key if they decided not to confess. The other half experienced the opposite mapping. After participants had indicated their choice, they were shown a fictive feedback indicating the
opponent’s choice. The order of presentation of the 8 photographs was randomized and each photo remained visible until each game ended.
Instructions explicitly defined the decision to confess as a competitive behavior and the choice to deny as a cooperative behavior. Instructions required participants to associate each opponent’s face to his/her behavior (cooperative or competitive). To facilitate the retrieval of the information about the opponent’s face and his/her behavior, two different versions of the social dilemma procedure were created. In version 1, female opponents always exhibited a cooperative behavior, while male opponents exhibited a competitive behavior. In version 2 the association gender-behavior was reversed. Half of the participants were administered version 1, while the other half were administered version 2.
To verify whether participants memorized the association between the opponent’s face and his/her behavior, at the end of the social dilemma game they were administered a true/false test in which each of the 8 faces was associated with a statement that could be either true or false (e.g.,: He/she chose to cooperate with/compete against you). For each face, four statements (two true and two false) were presented for a total of 32 trials. The order of presentation of the trials was
randomized. Participants responded by pressing the V-key on the keyboard if the statement was true, and the F-key if the statement was false. If the participant gave a wrong answer, the whole test was administered again until participant’s accuracy reached 100%.
Once participants successfully completed the true/false questionnaire, they were required to rate the social interaction with each opponent using a 6-point bipolar semantic differential scale on the following dimensions (Iani et al., 2011): easy vs. difficult (1 = difficult, 6 = easy), pleasant vs. unpleasant (1 = unpleasant, 6 = pleasant), positive vs. negative (1 = negative, 6 = positive), cooperative vs. highly-cooperative (1 = cooperative, 6 = highly-cooperative) and non-competitive vs. highly-non-competitive (1 = non-non-competitive, 6 = highly-non-competitive). Immediately after, they took part in a gaze-cuing task in which the same 8 faces presented in the social dilemma task were used as cuing faces.
Figure 1. Illustration of the gaze-cuing procedure depicting examples of stimuli (not drawn to scale) and sequence of events for a spatially congruent trial (top) and a spatially incongruent trial
Each trial began with the presentation of a white fixation cross (0.83°) in the center of a black screen for 900 ms (fixation frame), followed by a central face with straight gaze (face frame). After 900 ms, the same face appeared with averted gaze (cue frame). After 200 ms, a white target letter (L or T, 0.83°) appeared 10.8° to the left or right of the center of the screen in one of two possible locations that could be spatially congruent or incongruent with gaze direction. The target frame remained visible until a response was provided (see Dalmaso et al., 2012, for a similar gaze-cuing procedure). Gaze direction was uninformative as regards target location. Participants were explicitly instructed to respond according to the letter identity while ignoring the distracting face and to maintain fixation at the center of the screen. We chose a 200-ms stimulus onset asynchrony (SOA) between gaze-cue and target because this interval is long enough to allow the encoding of the gaze and, at the same time, it is short enough to automatically cue attention (see Frischen, Bayliss, & Tipper, 2007 for a review). Furthermore, this SOA has also been used in previous studies assessing the modulation of gaze-cuing by social factors (Dalmaso et al., 2012; Pavan et al., 2011) hence we chose it to facilitate comparisons.
Half of the participants were instructed to press the D-key on the keyboard if the target was an ‘L’, and the K-key if the target was a ‘T’. The remaining participants responded using the opposite stimulus-response mapping. Instructions emphasized both response speed and accuracy. No feedback was provided.
There were 64 trials for each combination of congruency between gaze direction and target location (congruent vs. incongruent) and social relation with the previous social dilemma’s
opponent who now served as distracter. A total of 256 trials were randomly presented and divided into two equal blocks of 128 trials each. A short rest was allowed between blocks. A practice session of 16 trials was given prior to the beginning of the first block. At the end of the
experimental session, participants were asked if they had guessed the main aim of the experiment. All participants answered “no” and claimed that they believed to the information provided by the experimenter at the beginning of the session.
2.2 RESULTS AND DISCUSSION
2.2.1 Social dilemma
Overall, 61.2% of the participants decided to cooperate. A one sample t test showed that this percentage marginally differed from the chance level of 50%, t(23) = 1.98, p = .06. Since both male and female photos were used, we computed the percentage of participants who chose to cooperate or compete as a function of the opponent’s gender. Cooperative choices did not differ from the chance level both when the face belonged to a female (61.5%, t(23) = 1.85, p = .08) or it belonged to a male (60.4%, t(23) = 1.59, p = .13). This result indicates that the gender of the face did not affect participants’ performance at the social dilemma.
2.2.2 Subjective Ratings
T tests for independent samples were used to compare the subjective ratings of the social dilemma situation involving cooperative and competitive opponents. Scores significantly differed between cooperative and competitive opponents for all dimensions (all ps< .001) (see Table 1).
The finding that when the opponent displayed a competitive behavior the situation was judged as more competitive and as less cooperative than when he/she displayed a cooperative behavior confirmed that our manipulation was effective.
Social Relationship Cooperative Competitive t p-value Cooperative 5.2 (0.7) 1.8 (0.7) 12.3 < .001 Competitive 2.2 (1.1) 4.9 (1.0) -6.7 < .001 Easy 4.8 (0.9) 2.7 (0.8) 8.7 < .001 Pleasant 4.7 (0.7) 2.1 (0.7) 10.0 < .001
Positive 4.9 (0.9) 2.0 (0.7) 10.0 < .001
Table 1. Experiment 1: Mean subjective ratings (and standard deviations) of the social interaction with the opponent as a function of the opponent’s behavior (cooperative vs. competitive).
2.2.3 Gaze-cuing Responses
Practice trials, errors (2.9% of the total trials) and trials with RTs faster than 150 ms or slower than 3000 ms (0.2% of the total trials) were excluded from the analyses.
Correct mean RTs were entered into a repeated-measures analysis of variance (ANOVA) with cue-target spatial congruency (congruent vs. incongruent) and social relationship (cooperative vs. competitive) as within-participant factors and version of the social dilemma (version 1: cooperative female opponents and competitive male opponents; version 2: cooperative male opponents and competitive female opponents) as between-participants factor. When necessary, comparisons were performed using paired samples t tests and by correcting the p value for the number of comparisons (Bonferroni correction).
The analysis revealed a significant main effect of cue-target spatial congruency, F(1,22) = 14.67, p = .001, ηp2 = .40, indicating that participants were faster to identify the target when it
appeared in the gazed-at location (i.e., congruent trials, M = 586 ms, SE = 16.81 ms, 95% CI from 551 to 621 ms) as compared to when it appeared in the opposite location (i.e., incongruent trials, M = 603 ms, SE = 17.43, 95% CI from 567 to 639 ms). The main effect of social relationship did not reach significance, F < 1, but it interacted with cue-target spatial congruency, F(1,22) = 8.98, p < . 05, ηp2 = .29. Post-hoc showed that the difference between congruent and incongruent trials (i.e.,
gaze-cuing effect) was significant (26 ms, 95% CI from 15 to 38 ms) for competitive opponents (581 vs. 607 ms, t(23) = 4.64, p Bonferroni corrected < .001), but did not reach significance (8 ms, 95% CI from -2 to 19 ms) for cooperative opponents (591 vs. 599 ms, t(23) = 1.74, p Bonferroni corrected = .38) (see Figure 2). Neither congruent nor incongruent trials differ across the two social relationship
conditions (ps Bonferroni corrected > .33). The main effect of version did not reach significance, F(1,22) = 2.84, p =.11, and did not interact with the other factors, all Fs < 1.
Figure 2. Mean RTs (± s.e.m.) in ms, collapsed across versions, for spatially congruent (grey) and incongruent (black) trials as a function of the social relation with the cuing face for Experiment 1
(panel a) and Experiment 2 (panel b). Asterisks denote significant differences.
In order to evaluate the time course of the influence of the social manipulation on the gaze-cuing effect, a distributional analysis of RTs was performed (Ratcliff, 1979). To have enough observations in each bin, we chose to divide the RT distribution in quartiles. Thus, individual correct RTs for each condition were rank ordered and divided into four bins. Mean RTs for each bin were then entered into a repeated-measures analysis of variance (ANOVA) with cue-target spatial congruency (congruent vs. incongruent), social relationship (cooperative vs. competitive) and bin (1 to 4) as within-participant factors.
Besides the effects and interactions already discussed in the overall analysis, the analysis showed a main effect of bin, F(3,69) = 184.78, p < .001, ηp2 = .89, that reflected RTs increasing with
bin. The two-way interaction between bin and cue-target spatial congruency was significant,
F(3,69) = 3.43, p < .05, ηp2 = .13, as was the three-way interaction between bin, cue-target spatial
congruency, and social relationship, F(3,69) = 3.74, p < .05, ηp2 = .14 (see Figure 3). Post-hoc
showed that when the distracter face belonged to a previously cooperative opponent the difference between congruent and incongruent trials was significant only in the first bin, t(23) = 3.30, p Bonferroni corrected = .02 (13, 8, 4, 10 ms from bin 1 to bin 4). Differently, when the face belonged to a
previously competitive opponent the difference was significant in the first bin, t(23) = 3.67, p Bonferroni corrected = .01, and in the last bin, t(23) = 3.18, p Bonferroni corrected = .03. The difference was marginally significant in the second, t(23) = 2.93, p Bonferroni corrected = . 06, and third bin, t(23) = 2.85, p Bonferroni corrected = . 07 (17, 12, 14, 60 ms from bin 1 to bin 4).
0 10 20 30 40 50 60 70 G C E ( m s)
Figure 3. Experiment 1: Magnitude of the gaze-cuing effect (± s.e.m.) in ms for the competitive (dotted line) and cooperative (solid line) social relation conditions as a function of bin.
Our results indicate that the type of relationship (i.e., cooperative or competitive) established during a previous interaction with a cuing face can modulate the gaze-cuing effect in female partici-pants characterized by an average level of competition, who oriented their attention in the same gaze direction of faces that during the social dilemma were associated with a competitive behavior, irrespective of their gender. The distributional analysis showed that the gaze-cuing effect was present for the fastest responses for both cooperative and competitive opponents. However, for competitive opponents only, the effect increased for longer responses. This pattern of results sug-gests that the type of prior interaction with the seen face modulates joint attention only at a later stage, probably only after information associated to the seen face has been retrieved from memory.
3. EXPERIMENT 2
The results of Experiment 1 indicate that females characterized by an average level of
competition are more likely to orient their attention in the same gaze direction of faces that during a prior interaction were associated with a competitive behavior. Since only average competitive females were tested, it is not possible to assess whether this result represents a consistent behavior within the female population or rather is influenced by the competitiveness level displayed by participants, as occurs with individual differences in self-esteem (Willowski, Robinson, & Friesen, 2009), anxiety (Putman, Hermans, & Van Honk, 2006) and introversion/extraversion traits (Ponari, Trojano, Grossi, & Conson, 2013).
Moreover, it is possible that in Experiment 1 the cooperation/competition manipulation created a dominance hierarchy, in which faces associated with competitive outcomes were
identified as more dominant. Hence the effects observed could be due to perceived dominance of a face rather than to the competitive behavior displayed during a previous interaction. Indeed,
dominance has been shown to affect gaze-cuing both in non-human primates (Shepherd, Deaner, & Platt, 2006) and in humans (Jones et al., 2010). For instance, Shepherd et al. (2006) showed that high-status monkeys were significantly more likely to follow the gaze of other high-status monkeys than low-status monkeys, while the low-status monkeys tended to follow the gaze of all the other monkeys.
The present experiment was aimed at testing these alternative interpretations. To this end, we replicated Experiment 1 with two groups of females classified as highly or lowly competitive according to the Competitiveness Index questionnaire (Smither & Houston, 1992; Houston et al., 2000). If females tend to always follow the gaze of individuals associated to a competitive behavior, then the gaze-cuing effect for competitive contenders should be present irrespective of the
individual level of competitiveness displayed by the participants. However, given the pivotal role that gaze-triggered orienting of attention plays in social monitoring (e.g., Emery, 2000), individuals who differ in the level of competitiveness may differently monitor social environment in
goal-directed tasks. More precisely, lowly competitive females, similarly to average competitive ones, should be more likely to orient their attention in the cued direction only for faces of competitors, since they may represent a potential obstacle to the achievement of their goal. On the contrary, highly competitive individuals should be more likely to follow the gaze of others irrespective of the outcomes, given that they tend to perceive others as potential rivals.
Differently, if the results of Experiment 1 are due to the fact that faces associated to a competitive behavior are perceived as more dominant, then, in line with previous results (e.g. Shepherd et al., 2006), lowly competitive individuals should be more likely to orient their attention in the cued direction irrespective of the outcomes of a prior interaction with a cueing face since due to their low competitiveness are likely to perceive others in general always as more dominant. As the level of individual competitiveness increases, individuals should be more likely to follow only the gaze of those faces associated to a competitive behavior.
3.1 METHODS
3.1.1 Participants
Thirty-two new undergraduate females (3 left-handed; M = 19.53 years, SD = 1.46 years) from the University of Modena and Reggio Emilia received course credit to participate to the study. All had normal or corrected-to normal vision and were naïve as to purpose of the experiment. The study was conducted in accordance with the ethical standards laid down in the Declaration of Hel-sinki, and fulfilled the ethical standard procedure recommended by the Italian Association of Psy-chology (AIP). Written consent was obtained for all of them and they were debriefed about the study at the end of the experiment.
They were selected from a larger sample (n=110) based on their scores at the Competitiveness Index (CI) questionnaire (Smither & Houston, 1992; Houston et al., 2000). To be included in the study, they had to score either in the high (12-20) or in the low (0-4) competitiveness range. Mean
scores were 15.56 (SD = 2.53) for the highly competitive group (n = 16) and 3.00 (SD = 1.26) for the lowly competitive group (n = 16).
3.1.2 Stimulus, apparatus and procedure
Stimuli, apparatus and procedure were the same as those used in Experiment 1.
3.2 RESULTS AND DISCUSSION
3.2.1 Social dilemma
Overall, 46.1% of highly competitive participants and 56.3% of lowly competitive participants decided to cooperate. A one sample t test (two-tailed) showed that these percentages did not differ from the chance level of 50% (High: t(15) = .58, p = .57; Low: t(15)= .91, p = .38). The
cooperative choices of highly competitive participants did not differ from the chance level either when the face belonged to a female (46.9%, t(15) = 0.42 p = .68) or when it belonged to a male (60.9%, t(15) = 1.28, p = .22). Lowly competitive participants chose to cooperate more when the face belonged to a female (51.6%) than when it belonged to a male (60.9%). This latter percentage did not differ from the chance level of 50% (female face: t(15) = -0.19, p = .86; male face: t(15) = 1.28, p = .22). As for Experiment 1, this result suggests that the gender of the face did not affect participants’ performance at the social dilemma.
3.2.2 Subjective Ratings
T tests for independent samples were used to compare the subjective ratings of the social dilemma situation involving cooperative and competitive opponents. The scores for both highly and lowly competitive groups significantly differed between cooperative and competitive opponents for all dimensions (all ps< .001) (see Table 2).
The finding that the situation was judged as more competitive when the opponent displayed a competitive behavior and as more cooperative when he/she displayed a cooperative behavior clearly indicates that our manipulation was effective.
Social Relationship Cooperative Competitiv e T p-value Cooperative Highly 5.9 (0.3) 1.2 (0.4) 33.3 < .001 Lowly 5.9 (0.3) 1.1 (0.3) 37.0 < .001 Competitive Highly 1.2 (0.5) 5.2 (1.0) 12.0 < .001 Lowly 1.2 (0.6) 5.3 (0.7) 13.9 < .001 Easy Highly 4.9 (1.1) 2.8 (1.5) 3.8 = .002 Lowly 4.8 (0.8) 2.4 (1.0) 6.31 < .001 Pleasant Highly 5.1 (1.1) 2.0 (1.3) 5.6 < .001 Lowly 4.9 (0.9) 1.7 (1.0) 8.93 < .001 Positive Highly 5.3 (1.1) 1.9 (1.3) 5.9 < .001 Lowly 5.0 (0.7) 1.7 (1.0) 8.51 < .001
Table 2. Experiment 2: Mean subjective ratings (and standard deviations) of the social interaction with the opponent as a function of the opponent’s behavior (cooperative vs. competitive) for highly
and lowly competitive participants.
3.2.3 Gaze-cuing Responses
Practice trials, errors (3.5% of the total trials) and trials with RTs faster than 150 ms or slower than 3000 ms (0.1% of the total trials) were excluded from the analyses.
Correct mean RTs were entered into a repeated-measures analyses of variance (ANOVA) with cue-target spatial congruency (congruent vs. incongruent) and social relationship (cooperative vs. competitive) as within-participant factors and version of the social dilemma (version 1:
cooperative female opponents and competitive male opponents; version 2: cooperative male opponents and competitive female opponents) and level of competition (high vs. low) as between-participants factor. When necessary, comparisons were performed using paired samples t tests and by correcting the p value for the number of comparisons (Bonferroni correction).
The analysis revealed a main effect of social relationship, F(1,28) = 9.06, p = .005, ηp2 = .14,
comparisons revealed that participants who performed version 2 of the social dilemma responded faster to cooperative (M = 520 ms, SE = 15.50 ms, 95% CI from 489 to 552 ms) than competitive opponents (M = 535 ms, SE = 17.17 ms, 95% CI from 500 to 571 ms). Differently, RTs for cooperative and competitive opponents did not differ for participants who performed version 1 (cooperative opponents: M = 554 ms, SE = 15.51 ms, 95% CI from 522 to 586 ms; competitive opponents: M = 551ms, SE = 17.17 ms, 95% CI from 516 to 587 ms). This result indicates that responses on the gaze-cuing task were faster when the distractor’s face belonged to a cooperative male. This finding is in line with studies on face recognition and memory (e.g., Maleay, 1996; Oda, 1999) which reported that faces of cooperative males are recognized less than when the same males are displayed as competitive individuals. Thus, the overall faster performance in version 2 can be explained by assuming that cooperative faces (male faces) interfered less with task performance.
The main effect of cue-target spatial congruency was significant, F(1,28) = 12.78, p = .001,
ηp2 = .310, indicating that participants were faster to identify the target when it appeared in the
gazed-at location (i.e., congruent trials, M = 534 ms, SE = 11.41 ms, 95% CI from 511 to 558 ms) as compared to when it appeared in the opposite location (i.e., incongruent trials, M = 546 ms, SE = 11.78, 95% CI from 522 to 570 ms).
The three-way interaction between social interaction, cue-target congruency, and level of competition was significant, F(1,28) = 4.72, p = .0538, ηp2 = .14. Separate analyses by the level of
competitiveness showed that the interaction between cue-target congruency and social interaction was significant for the lowly competitive group, F(1,28) = 5.90, p < .05, ηp2 = .30. Post-hoc tests
showed that the difference between congruent and incongruent trials (i.e., gaze-cuing effect) was significant (27 ms, 95% CI from 8 to 45ms) for competitive opponents (565 vs. 538ms, t(15) = 3.12, p Bonferroni corrected < .028), while no effect was evident (-6 ms, 95% CI from -22 to 9 ms) for cooperative opponents (542 vs. 548 ms, t(15) = -.86, p Bonferroni corrected = 1.61) (see Figure 2). For the highly competitive group, there was only a main effect of cue-target congruency, F(1,28) = 6.33, p
< .05, ηp2 = .31, with faster RTs in congruent (525 ms) than in incongruent (539 ms) trials. The
interaction between cue-target congruency and social interaction did not reach significance, F < 1. These results indicate that the cooperative and competitive outcomes of a previous interaction with a cuing face can modulate the gaze-cuing effect in female participants characterized by a low level of competition, who oriented their attention in the same gaze direction of faces that during the social dilemma were associated with a competitive behavior. Differently, the outcomes of a prior interaction did not influence the gaze triggered orienting of attention in females characterized by a high level of competition.
4. GENERAL DISCUSSION
The results of the present study demonstrate that the type of relationship (i.e., cooperative or competitive) established during a previous interaction with a cuing face can modulate the gaze-cuing effect in female participants characterized by low and average levels of competition, who oriented their attention in the same gaze direction of faces that during the social dilemma were associated with a competitive behavior, irrespective of their gender. Differently, females
characterized by a high level of competition were not affected by a prior cooperative or competitive interaction with the cueing face. This result allows us to exclude that the results of Experiment 1 were due to the cooperation/competition manipulation leading to the emergence of a dominance hierarchy. Furthermore, they suggest that individuals characterized by different levels of
competitiveness may differently monitor the social environment when engaged in goal-directed tasks.
The results of the present study are in line with prior studies suggesting that joint attention is not immune to social modulations but they extend these results in two novel ways. First, they show that, not only specific identity cues of an observed face, such as social status (Dalmaso et al., 2012), political affiliation (Liuzza et al., 2011), race (Pavan et al., 2011), and age (Ciardo et al., 2014), but also whether he/she decided to cooperate or compete during a prior interaction can modulate the
gaze-cuing effect. Interestingly, this result confirms the carry-over effect of competition found in previous studies (e.g., Iani et al., 2014; Sessenberg et al., 2007) since the effects of competition introduced in the social dilemma game were still evident and affected performance in the
subsequent gaze-cuing task. Second, they show that there is a strong tendency to orient attention in the same direction of the gaze of those individuals that may represent an important source of information and may be relevant for the attainment of our goals, even though these individuals may be categorized as out-group members, and irrespective of their gender.
In competitive situations it may be fundamental for the observer to orient attention to others’ actions in order to monitor their progress toward their goals (Poortvliet & Darnon, 2010). Following a competitor’s gaze may represent an highly adaptive behavior, as it allows a constant monitoring of his/her intentions and, consequently, it allows the observer to plan the more adaptive behavior to maximize the chance of achieving his/her goals. It is therefore possible that the type of relationship established with others biases attention toward those individuals that are more salient in the social environment (Ristic & Kingstone, 2005) because of the role that they could play in favoring or hindering the attainment of our personal goals. In support of this line of thoughts are the results of a series of studies showing that pictures of non-cooperative individuals attract more automatic
attention than pictures of cooperative individuals (Vanneste et al., 2007) and that faces of cheaters and partners are different encoded into memory, with those of cheaters being memorized more accurately than those of non-cheaters (e.g., Chiappe et al., 2004; Mealy, Daood, & Krage, 1996; Oda, 1997; Yamagishi et al., 2003), possibly due to beliefs of trustworthy and personality traits that we implicitly associate to them (Bayliss & Tipper, 2006). Our results also support the proposal that the gaze-triggered orienting is augmented under conditions characterized by a low relational value (Wilkowski, Robinson and Friesen, 2009), thus pointing up the pivotal role that gaze-triggered orienting of attention plays in social monitoring (Pickett & Gardner, 2005).
Since we tested only female participants, we do not know whether our results may extend to males. It is however possible that they do not generalize given the reported gender differences in
sensitivity to social cues (e.g., Bayliss et al., 2005; Geary, 2010 for a review) and in emotional responses to cooperation and competition (e.g., Kivikangas, Kätsyri, Järvelä, & Ravaja, 2014). Future target studies should address this issue.
To conclude, the present study shows, for the first time, that different types of social interactions may modulate joint attention differently. The present results shed light on the role of competition in modulating gaze-mediated attention, suggesting that during social interactions more attentional resources are allocated to those individuals with greater social relevance and from whom we can draw more information to achieve our goals.
Funding
REFERENCES
Balota, D.A., Yap, M.J. (2011). Moving beyond the mean in studies of mental chronometry: The power of response time distributional analyses. Current Directions in Psychological
Science, 20, 160-166. doi: 0.1177/0963721411408885
Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind. Cambridge, MA: MIT Press.
Bayliss, A. P., & Tipper, S. P. (2006). Predictive Gaze Cues and Personality Judgments Should Eye Trust You? Psychological Science, 17, 514-520. doi:10.1111/j.1467-9280.2006.01737 Bayliss, A. P., Di Pellegrino, G. D., & Tipper, S. P. (2005). Sex differences in eye gaze and
symbolic cueing of attention. The Quarterly Journal of Experimental Psychology, 58, 631-650. doi:10.1080/02724980443000124
Bayliss, A.P., Schuch, S., Tipper, S.P. (2010). Gaze cuing elicited by emotional faces is influenced by affective context. Visual Cognition, 18, 1214-1232. doi:10.1080/13506285.2010.484657 Bonifacci, P., Ricciardelli, P., Lugli, L., & Pellicano, A. (2008). Emotional attention: effects of
emotion and gaze direction on overt orienting of visual attention. Cognitive Processing, 9, 127-135.
Chan, E., & Ybarra, O. (2002). Interaction goals and social information processing:
Underestimating one's partners but overestimating one's opponents. Social Cognition, 20, 409-439. doi: 10.1521/soco.20.5.409.21126
Chiappe, D., Brown, A., Dow, B., Koontz, J., Rodriguez, M., & McCulloch, K. (2004). Cheaters are looked at longer and remembered better than cooperators in social exchange situations.
Evolutionary Psychology, 2, 108-120.
Ciardo, F., Marino, B. F., Actis-Grosso, R., Rossetti, A., & Ricciardelli, P. (2014). Face age modulates gaze following in young adults. Scientific Reports, 4, 4746.
Dalmaso, M., Galfano, G., Coricelli, C., & Castelli, L. (2014). Temporal dynamics underlying the modulation of social status on social attention. PLoS ONE, 9, e93139.
doi:10.1371/journal.pone.0093139
Dalmaso, M., Pavan, G., Castelli, L., & Galfano, G. (2012). Social status gates social attention in humans. Biology Letters, 16, 450-452. doi:10.1098/rsbl.2011.0881
De Jong, R., Liang, C.-C., & Lauber, E. (1994). Conditional and unconditional automaticity: A dual-process model of effects of spatial stimulus-response correspondence. Journal of
Experimental Psychology: Human Perception and Performance, 20, 731-750.
Deaner, R. O., Shepherd, S. V., & Platt, M. L. (2007). Familiarity accentuates gaze cuing in women but not men. Biology Letters, 3, 65-68. doi:10.1098/rsbl.2006.0564
Driver, J., Davis, G., Ricciardelli, P., Kidd, P., Maxwell, E., & Baron-Cohen, S. (1999). Gaze Perception Triggers Reflexive Visuospatial Orienting. Visual Cognition, 6, 509-540. doi:10.1080/135062899394920
Emery, N. J. (2000). The eyes have it: the neuroethology, function and evolution of social gaze.
Neuroscience & Biobehavioral Reviews, 24, 581-604.
Frischen, A., Bayliss, A. P., & Tipper, S. P. (2007). Gaze cueing of attention: Visual attention, social cognition, and individual differences. Psychological Bulletin, 133, 694-724. doi:10.1037/0033-2909.133.4.694
Galfano, G., Dalmaso, M., Marzoli, D., Pavan, G., Coricelli, C., & Castelli, L. (2012). Eye gaze cannot be ignored (but neither can arrows). The Quarterly Journal of Experimental
Psychology, 65, 1895-1910. doi: 10.1080/17470218.2012.663765
Geary, D. C. (2010). Male, female: The evolution of human sex differences (2nd edition). Washington, DC: American Psychological Association.
Greene, D.J., Mooshagian, E., Kaplan, J.T., Zaidel, E., Iacoboni, M. (2009). The neural correlates of social attention: automatic orienting to social and nonsocial cues. Psychological
Hayward, D. A., & Ristic, J. (2013). The uniqueness of social attention revisited: working memory load interferes with endogenous but not social orienting. Experimental Brain Research, 231, 405-414. doi: 10.1007/s00221-013-3705-z
Houston, J. M., Kinnie, J., Lupo, B., Terry, C., & Ho, S. S. (2000). Competitiveness and conflict behavior in simulation of a social dilemma. Psychological Reports, 86, 1219-1225.
Iani, C., Anelli, F., Nicoletti, R., Arcuri, L., & Rubichi, S. (2011). The role of group membership on the modulation of joint action. Experimental Brain Research, 211, 439-445.
doi:10.1007/s00221-011-2651-x
Iani, C., Anelli, F., Nicoletti, R., Rubichi, S. (2014). The carry-over effect of competition in task-sharing: Evidence from the joint Simon task. PLoS ONE, 9,
e97991.doi:10.1371/journal.pone.0097991
Innocenti, A., De Stefani, E., Bernardi, N. F., Campione, G. C., & Gentilucci, M. (2012). Gaze direction and request gesture in social interactions. PLoS ONE, 7, e36390.doi:
10.1371/journal.pone.00363900.
Jones, B. C., DeBruine, L. M., Main, J. C., Little, A. C., Welling, L. L., Feinberg, D. R., &
Tiddeman, B. P. (2009). Facial cues of dominance modulate the short-term gaze-cuing effect in human observers. Proceedings of the Royal Society B: Biological Sciences,
rspb20091575.
Kivikangas, J.M., Kätsyri, J., Järvelä, S., & Ravaja, N. (2014). Gender differences in emotional responses to cooperative and competitive game play. PLoS ONE, 9, e100318.
doi:10.1371/journal.pone.0100318
Langton, S. R., Watt, R. J., & Bruce, V. (2000). Do the eyes have it? Cues to the direction of social attention. Trends in Cognitive Sciences, 4, 50-59.
Liuzza, M. T., Cazzato, V., Vecchione, M., Crostella, F., Caprara, G. V., & Aglioti, S. M. (2011). Follow my eyes: the gaze of politicians reflexively captures the gaze of ingroup voters.
Mealy, L., Daood, C., & Krage, M. (1996). Enhanced memory for faces of cheaters. Ethology and
Sociobiology, 17, 119 – 128.
Minear, M., & Park, D. C. (2004). A lifespan database of adult facial stimuli. Behavior Research
Methods, Instruments, & Computers, 36, 630-633.
Oda, R. (1997). Biased face recognition in the prisoner's dilemma game. Evolution and Human
Behavior, 18, 309-315. doi:10.1016/S1090-5138(97)00014-7
Ohlsen, G., van Zoest, W. & van Vugt, M. (2013). Gender and facial dominance in gaze cuing: emotional context matters in the eyes that we follow. PLoS ONE, 8, e59471.
doi:10.1371/journal.pone.0059471
Pavan, G., Dalmaso, M., Galfano, G., & Castelli, L. (2011). Racial Group Membership Is Associated to Gaze-Mediated Orienting in Italy. PLoS ONE, 6, e25608.
doi:10.1371/journal.pone.0025608
Pickett, C.L., & Gardner, W.L. (2005). The social monitoring system: Enhanced sensitivity to social cues and information as an adaptive response to social exclusion and belonging need. In K. Williams, J. Forgas, & W. von Hippel (Eds.), The social outcast: Ostracism, social
exclusion, rejection, and bullying (pp. 213–226). New York: Psychology Press.
Pierno, A. C., Becchio, C., Wall M. B., Smith A. T., Turella, L., & Castiello, U. (2006). When gaze turns into grasp. Journal of Cognitive Neuroscience, 18, 2130-2137.
Ponari, M., Trojano, L., Grossi, D., & Conson, M. (2013). “Avoiding or approaching eyes”? Introversion/extraversion affects the gaze-cueing effect. Cognitive Processing, 14, 293-299. Poortvliet, P. M., & Darnon, C. (2010). Toward a more social understanding of achievement goals.
The interpersonal effects of mastery and performance goals. Current Directions in
Psychological Science, 19, 324-328.doi:10.1177/0963721410383246
Poundstone, W. (1992). Prisoner's dilemma: John von Neumann, game theory, and the puzzle of
Putman, P., Hermans, E., & Van Honk, J. (2006). Anxiety meets fear in perception of dynamic expressive gaze. Emotion, 6, 94–102.
Rabbie, J. M., & Horwitz, M. (1969). Arousal of ingroup-outgroup bias by a chance win or loss. Journal of Personality and Social Psychology, 13, 269-277.
Ratcliff, R. (1979). Group reaction time distributions and an analysis of distribution statistics.
Psychological Bulletin, 86, 446-461.
Ricciardelli, P., Carcagno, S., Vallar, G., & Bricolo, E. (2013). Is gaze following purely reflexive or goal-directed instead? Revisiting the automaticity of orienting attention by gaze cues.
Experimental Brain Research, 224, 93-106. doi:10.1007/s00221-012-3291-5
Ristic, J., & Kingstone, A. (2005). Taking control of reflexive social attention. Cognition, 94, B55-B65. doi: 10.1016/j.cognition.2004.04.005
Sassenberg, K., Moskowitz, G. B., Jacoby, J., & Hansen, N. (2007). The carry-over effect of competition: The impact of competition on prejudice towards uninvolved outgroups.
Journal of Experimental Social Psychology, 43, 529-538. doi:10.1016/j.jesp.2006.05.009
Shepherd, S. V., Deaner, R. O., & Platt, M. L. (2006). Social status gates social attention in monkeys. Current Biology, 16, R119-R120.
Smither, R. D., & Houston, J. M. (1992). The nature of competitiveness: The development and validation of the Competitiveness Index. Educational and Psychological Measurement, 52, 407-418.
Toma, C., Yzerbyt, V., & Corneille, O. (2010). Anticipated cooperation vs. competition moderates interpersonal projection. Journal of Experimental Social Psychology, 46, 375-381.
doi:10.1016/j.jesp.2009.11.005
Vanneste, S., Verplaetse, J., Van Hiel, A., & Braeckman, J. (2007). Attention bias toward
noncooperative people. A dot probe classification study in cheating detection. Evolution and
Vonk, R. (1998). Effects of cooperative and competitive outcome dependency on attention and impression preferences. Journal of Experimental Social Psychology, 34, 265-288.
Yamagishi, T., Tanida, S., Mashima, R., Shimoma, E., & Kanazawa, S. (2003). You can judge a book by its cover: Evidence that cheaters may look different from cooperators. Evolution
and Human Behavior, 24, 290-301. doi:10.1016/S1090-5138(03)00035-7
Wilkowski, B. M., Robinson, M. D., & Friesen, C. K. (2009). Gaze-triggered orienting as a tool of the belongingness self-regulation system. Psychological Science, 20, 495-501.
Table and Figure captions
Table 1. Experiment 1: Mean subjective ratings (and standard deviations) of the social interaction with the opponent as a function of the opponent’s behavior (cooperative vs. competitive).
Table 2. Experiment 2: Mean subjective ratings (and standard deviations) of the social interaction with the opponent as a function of the opponent’s behavior (cooperative vs. competitive) for highly and lowly competitive participants.
Figure 1. Illustration of the gaze-cuing procedure depicting examples of stimuli (not drawn to scale) and sequence of events for a spatially congruent trial (top) and a spatially incongruent trial
(bottom).
Figure 2. Mean RTs (± s.e.m.) in ms, collapsed across versions, for spatially congruent (grey) and incongruent (black) trials as a function of the social relation with the cuing face for Experiment 1 (panel a) and Experiment 2 (panel b). Asterisks denote significant differences.
Figure 3. Experiment 1: Magnitude of the gaze-cuing effect (± s.e.m.) in ms for the competitive (dotted line) and cooperative (solid line) social relation conditions as a function of bin.
Table 1 Social Relationship Cooperative Competitive t p-value Cooperative 5.2 (0.7) 1.8 (0.7) 12.3 < .001 Competitive 2.2 (1.1) 4.9 (1.0) -6.7 < .001 Easy 4.8 (0.9) 2.7 (0.8) 8.7 < .001 Pleasant 4.7 (0.7) 2.1 (0.7) 10.0 < .001 Positive 4.9 (0.9) 2.0 (0.7) 10.0 < .001
Table 2 Social Relationship Cooperative Competitiv e t p-value Cooperative Highly 5.9 (0.3) 1.2 (0.4) 33.3 < .001 Lowly 5.9 (0.3) 1.1 (0.3) 37.0 < .001 Competitive Highly 1.2 (0.5) 5.2 (1.0) 12.0 < .001 Lowly 1.2 (0.6) 5.3 (0.7) 13.9 < .001 Easy Highly 4.9 (1.1) 2.8 (1.5) 3.8 = .002 Lowly 4.8 (0.8) 2.4 (1.0) 6.31 < .001 Pleasant Highly 5.1 (1.1) 2.0 (1.3) 5.6 < .001 Lowly 4.9 (0.9) 1.7 (1.0) 8.93 < .001 Positive Highly 5.3 (1.1) 1.9 (1.3) 5.9 < .001 Lowly 5.0 (0.7) 1.7 (1.0) 8.51 < .001
Figure 3 0 10 20 30 40 50 60 70 G C E ( m s)
