Continuous Theta Burst TMS over the
Posterior Superior Temporal Sulcus
might disrupt updating of Social Ties
By Andrew Sutjahjo
Student number 5873185
Supervisor: Nadége Bault
Word count: 6002
Social decisions are pervasive in our lives: from asking someone out on a date to negotiating a business deal. In these decisions that someone makes, the outcome is not only dependent on the choices they make, but also dependent on the choices of another person. Within this context it is not so strange to see that there are a plethora of mechanisms in the brain to notice and understand social information, predict what the other will do, and base one’s own choice on this prediction. In this paper, we will focus on the interplay between the domains of neuroscience and economics in researching cooperation and competition in an economic game. In particular, we have found a neural substrate that seems to lie at the basis of the behaviors shown in one of these games: the posterior Superior Temporal Sulcus (pSTS). When conducting an economic game while the participant was in an fMRI scanner, Bault et al. (in preparation) used the activation of the pSTS to estimate their social tie with their opponent, and was able to use that tie to predict the subsequent choice of the participant. In this paper, we will further flesh out the research towards the role of the pSTS by experimentally reducing its activity during the game with the technique of a continuous theta burst stimulation (cTBS) protocol of repetitive Transcranial Magnetic Stimulation (rTMS). But before we go into detail, some groundwork needs to be laid in social decision making.
Psychology has a term that is closely related to the understanding of others: Empathy. Empathy roughly refers to putting one’s self in another’s situation. According to Lamm (2007) this has three characteristics: (1) affective sharing, that is, an affective response to another person which may entail sharing that person’s (imagined) emotional state; (2) perspective taking, involving a cognitive capacity to take the other person’s perspective; and (3) cognitive appraisal, involving monitoring mechanisms that keep track of the origins (self vs. other) of the experienced feelings.
Simulation theory propounds that this is possible due to the recreation of the mental processes on ourselves that, if carried out, would reproduce the other’s behavior. One of the major discoveries of the past 20 years in psychology was that of mirror neurons. These are neurons that are both active during a particular action, as well as when another conspecific was seen performing the same action. For example when Di Pellegrino et al. (1992) recorded a single cell in the pre-motor cortex of a macaque monkey, it showed activation both when the monkey picked up a peanut with his hand, as well as when it watched another macaque or a human pick up a peanut.
Further research on humans has shown that increased scale on the empathy scale correlated with an increased activation with the mirror neuron network (Gazzola et al. 2006).
Although there is a lot of evidence that the mirror neuron system contributes to empathy, and possibly partially explains the prediction of others, the imitation of others cannot accurately predict what another person’s choice will be in a cognitively more taxing choice. Thus another system is needed to predict what other people will choose to do.
This cognitive capacity of inferring the mental representations and beliefs and desires of the other and being able to predict their actions is called Theory of Mind (ToM). In general this capacity has been attributed to the temperoparietal junction (TPJ). However, all of the studies used to find the involvement of the TPJ in theory of mind have used paradigms that made the participants actively mentalize what another person believes (Saxe 2006).
The understanding of the prediction of someone’s actions has been attributed to other areas. Several experiments have implicated the medial Prefrontal cortex (MPFC), from experiments of playing rock-paper-scissors with either a person or a computer (Gallagher et al. 2006), playing a trust game with a player or computer (McCabe et al 2001), and playing economic games with either a player or
computer (Rilling et al 2004). These experiments all observed a greater activity in the MPFC when playing with a player than playing with a computer. Rilling, however, also found that the pSTS was
Ramnani (2005) conducted a simple visual stimulus-response task; four boxes were shown and participants needed to react by pressing one of four buttons in response. Before each trial they were shown one of five cues, four of these indicated which box would have to be pressed when the time to respond came, and one indicated that the correct response would be shown together with the boxes. The cues were also given a color, indicating whether they were the ones that needed to respond, a person they were doing the experiment with, or a computer. No matter which color came they had to predict and monitor each action. They found that the pSTS was active during the prediction of the other, but not the computer.
Further studies support this role of the pSTS in the social perception of others, ranging from autism studies (Zilbovicius 2006); biological motion cues (Allison et al. 2000); whether or not someone would lie (Behrens 2009), stimuli that have acquired relevance during meaningful social interaction such as faces of cooperative game partners (Singer et al., 2004); keeping track of one’s influence on other agent’s intentions and strategies (Hampton et al., 2008; Haruno and Kawato, 2009)
How a person makes a choice in a social decision making task is a slightly more complicated case. Traditional economics has the idea of the Homo Oeconomicus, the notion that humans will always make a rational, earning maximizing decision when given the choice. While in many cases people do act this way, there are many examples that indicate otherwise. In many cases people show altruistic behavior, in which the decisions they make are ‘selfless’ and no personal gain is received.
In the trust game, 2 participants are playing with each other. One of the participants is given an endowment of points or money. He is then given the choice to give some or all of the money to the other player. The money that he gives to the other player is multiplied by a factor greater than one. If the other is given money, he then is allowed to choose whether or not he will give some of the money back to the first player. In a repeated game the player that is given the endowment switches with each trial of the game. Here it is found that players are both willing to share some of their endowment, as well as sharing back some of the money received after it is multiplied. This can still be explained by the theory of the Homo Oeconomicus. If all of the endowment is shared and shared equally by the second player, both players would earn more money by the end of the experiment. Thus it could be a strategic decision to share. Krueger et al. (2007) conducted this experiment in the fMRI and found that this has its neural correlate in the mPFC, where there is increased activity when people behaved altruistically at all times. This was interpreted as these people ‘mentalizing’ the other’s trusting intentions.
However, when the experiment was reproduced with only a single, anonymous interaction, it was found that this sharing behavior was still present. This shows that even when there is no strategic incentive to share, people are still willing to behave altruistically, although the prevalence of this behavior was smaller than in the repeated trials. (Berg et al. 1995)
In social decision making for altruism you could see that there are many brain areas involved. Moll et al. (2006) found that donating to charitable causes activated the same Ventral Tegmental area to striatum network that can be seen when monetary rewards are received. A different study by Tankersley et al. (2007) shows that the superior temporal cortex is needed in the perception of agency behind an action. They suggest that the superior temporal cortex allows an empathetic response to occur when only an action is seen, as the agent behind the action can be perceived due to the superior temporal cortex.
This willingness to behave altruistically regardless of previous interaction is usually described in the form of the social value orientation (SVO), which can be estimated by having subjects make choices between different monetary distributions regarding themselves and an unspecified other (Haruno and Frith 2010, Liebrand 1984). While the SVO towards unspecified others is considered not to
change over time, it has been seen that the SVO towards a single person with which someone has interacted can change depending on the nature of the interactions.
Van Dijk and van Winden, (1992, 1997) formalized this change in amounts of altruism due to multiple interactions in the social ties model, in which social ties are the attachment between individuals that build up during economic interactions. In this model, the social tie explains the willingness to behave altruistically due to the idea of interdependent utilities. It considers that the utility that an individual obtains from the welfare of the other, on top of the individual’s SVO is the social tie. Contrasting with other theories that propose a static additional weight above the SVO that might be attached to an individual’s welfare, the social tie is dynamic, evolves over time depending on the interaction history of the two persons and more importantly on the feelings the individual derived from the interaction. The social tie model is very appealing for it allows for various kinds of behaviors such as standard selfish behavior, standard rational behavior, strategic foresight, and fixed other-regarding preferences like altruism and inequity aversion. The model can describe (in its extended version) many commonly observed behaviors in prisoner dilemma types of games, including mimicking behavior and
reciprocity.
Another game that can be seen through the perspective of the SVO and social ties is the Public Good game. This game is played with two or more players. People can choose to contribute differing amounts into a public good. The money in the public good grows and is equally distributed between the players irrespective of whether or not they contributed. Similar to the Trust game, contributions were seen in one-shot games and during repeated games. (Ledyard 2005).
Fahrenfort et al. (2012) found that during a repeated Public Good game, previous cooperation success and future cooperation only correlated with pSTS activation, as well as that individual differences in pSTS activation was able to predict individual differences in pro-social investment behavior in subsequent rounds of the game. He does warn against over-interpreting these results: seeing as the pSTS is linked to so many basic operations of social cognition, it is highly probable that the pSTS is part of a highly complex network in which regions work together to achieve certain functions. This corresponds with Bault et al. (in preparation) in which the pSTS was dynamically encoding the kindness of the other’s contributions to the public good. Bault et al. also found a functional connectivity between the pSTS and the medial Prefrontal Cortex.
A variant of the public good game by Fehr & Gachter (2000) also emphasizes another facet of altruism. The game lends itself to the possibility of free riders, those that do not contribute to the common pot, but still reap the benefits. In this variant of the public good game, after the money from the common pot is distributed, players are given the opportunity to punish other players by paying a cost: they may hand in some of their earnings to take away (usually a greater amount of) earnings from another player. When this is done to a free rider, they tend to cooperate more in subsequent rounds. However, as this is also done in one-shot games and de Quervain et al. (2004) show that although cooperation is a consequence of altruistic punishment, the punishment in itself is seen to activate reward centers in the brain, thus it should still be classified as altruistic.
Not all punishments are linked to altruism, but can also simply involve the joy of being nasty. The simplest example of this is the experiment by Abbink & Sadrieh (2009): 2 players earned an endowment through filling in questionnaires, and afterwards were asked if they wanted to destroy part of the endowment of a random, anonymous other. 40% destroyed part of the other’s
endowment. This behavior doesn’t only occur when it has no cost, Zizzo and Oswald (2001) found that many people choose to reduce the earnings of others, even at their own cost. Furthermore, people prefer to quit bargaining when deals turn out to be disadvantageous (Guth et al., 1982, cited in Zizzo. 2008).
This joy of being nasty in economics is thought to coincide with competitiveness or status seeking behaviours. In these cases, people try to maximize the difference in payoff between another person and himself. To improve their status they can resort to lowering the payoff of the other person. Individuals sometimes engage in nasty behaviors such as sabotage (Harbring, et al. 2007), cheating to increase their status (Charness, et al.) or robbery. In an experiment, about 60% of participants took money previously earned by a peer when given the opportunity (Bosman and Van Winden. 2002). Although there have not been many experiments researching the neural correlates of status seeking behaviours, the few that have been done seem to implicate similar regions as positive interactions. Viewing a hated face activates the striatum, bilateral insula, and mPFC (Zeki et Romaya, 2008). The mPFC was also activated during gloating (when a participant wins a lottery while he receives
information that another participant chose the losing choice (Bault et al. 2008), they suggest that the interplay between reward and social reasoning mediates the competitive component of their lottery task.
The notion that a single brain area or hormone mediates both positive and negatively valanced emotions in a single system is not a strange one. In parochial (in- vs out-group) altruism the neuropeptide oxytocin modulates both in-group cooperation and (defensive) out-group non-cooperation (De Dreu 2012). Similarly, the pSTS could be a mediator in both non-cooperation and status seeking behaviours through the complex network of brain areas it is another part of.
To research this role of the pSTS as a mediator of the development of cooperation and status seeking, a task is needed in which both behaviours are possible and can be developed by the participant themselves, and is not cued to show a certain behavior. As far as we know, there are two experiments that have researched cooperation and competition together. However, they are inadequate to test this research question: Decety et al. (2004) devised an experiment in which the participant is cued whether to cooperate or compete with a confederate in a building task; De Bruijn et al. (2009) similarly devised a task in which the participant had to perform a motor task, or observe a confederate perform this task. Depending on the condition (cooperation or competition) He was rewarded or lost money when the other made a mistake. Thus new game had to be designed. The Public Good and Bad Game (PGBG) is a mix between the Public Good game (as described earlier) and the common pool game. The common pool game is a game in which participants can sequentially take resources from a common pool. Everything that is still in the common pool is multiplied by a factor greater than one. In the PGBG two players are given an equal endowment of tokens, and there is a common pool with a set amount of tokens. The players can simultaneously choose whether or not (and if so, how much) to contribute or take from this common pool. After this action the common pool is multiplied by a factor greater than one, and the contents of the common pool is divided equally between the two participants. After the tokens in the common pool are pocketed, the endowment and common pool are refilled and the participants play another round with each other. To provide the experimental condition, the activity of the pSTS will be dampened. Whereas fMRI can only prove correlations, by dampening activity, causal evidence can be accrued. Transcranial magnetic stimulation is a technique whereby a spool of quickly changing electric currents is placed on the skull, and induces an electrical pulse within the brain. When these pulses are administered in a cTBS protocol, the excitability of a target area can be lowered for around 40 minutes.
The cTBS protocol has been administered to the pSTS in 2 different studies: Bertini et al. (2010) administered cTBS to the right pSTS and found that it reduced multisensory integration of audio and visual stimuli; Van Kemenade (2012) used this protocol and found that the sensitivity to biological motion perception was reduced. Both found that the reaction times were also significantly reduced.
As the PGBG itself is being tested, we expect the PGBG to allow for the expression of both status seeking and cooperative behaviors, and that these are driven by the participant themselves. If the pSTS is involved in mediating status seeking behaviors, and not restricted to mediating
cooperative behaviors, we will expect that the cTBS protocol will influence the development of both the status seeking and cooperative behaviors. Bault found that there was a positive correlation between pSTS activity and the (positive) social tie in the PGG. Thus we expect that the dampening of the pSTS will decrease the social tie, and the pro-social behavior forming. In the status seeking behavior we expect that dampening the pSTS does not decrease the social tie further, but dampens the decrease of the social tie.
We will also expect that the differences between the SVO for a random person and the SVO of the interaction partner after the game will decrease in the cTBS condition as a reflection of the lower social tie formation during the PGBG.
Experimental procedures:
This experiment was a pilot study examining the effects of inhibiting the pSTS using cTBS on the PGBG. As there were only 3 participants, only descriptive statistics can be reported.
Subjects:
Subjects were 3 healthy volunteers (2 females) between the age of 23 and 27, who have had experience with repetitive Transcranial Magnetic Stimulation (rTMS). The local ethics committee of the department of Psychology of the University of Amsterdam approved the experiment and all subjects gave their written informed consent.
Tasks
The subjects were presented with the ring test of social value orientation (Liebrand, 1984) in which they were asked to choose between two “self-other” payoff combinations. These combinations allocate points to, or take away points from the decision maker (Self) and a randomly chosen other participant (Other) (see figure1). Choices are private, and the effects of the decision maker were only seen at the end of the experiment. These combinations are shown as two bar graphs representing the Self and Other’s payoffs. The values of these combinations lie on a circle with the origin as center with a radius of 500 points, with the horizontal axis representing the decision maker’s points, and the vertical axis representing the Other’s points. Choices can be seen as a vector from the origin of the circle. The participant makes 32 choices between 2 combinations, and by adding up vectors of these 32 choices, we find a measure of the participant’s Social Value Orientation (SVO).
Figure 1: Choices ring test
In the PGBG, there are two private accounts, one for each player, and a common account. At the start of every round, each private account is filled with 50 monetary units (MU) and the common account is filled with 100 MU. At the end of the round, each player receives the MU in their private account. In addition, both players receive 70% of the MU in the common account.
Each player may choose to do nothing, contribute MU to or take MU from the common account. Contributing and taking may be done in multiples of 10, up to 50 MU. However, each multiple of 10 given or taken carries a cost of 3 MU from their total earnings.
As you can see in equation 1: the payoff of player I at time t (Pit) is lowered by any taking or
contribution by himself (Cit), but is affected by the other player’s contribution (Cjt) in a larger way:
both players would be better off if both players would contribute equally (See appendix 1 and equation 1). After each round the participant is shown the other participant’s choice, and what each participant earned in that round.
P
it=120−
|
0.3 C
it|
+
0.7 C
jt(1)
This version of PGBG where the cost of contributing and taking is linear was chosen as previous pilot studies showed more competitive behavior in the linear-cost PGBG as opposed to the non-linear-cost PGBG.
The subjects were told they would subsequently be playing against two random participants, however they played 20 rounds of the PGBG against an simulation to control their opponent’s choices.. This simulation was set to model an altruistic player during the first game, and set to model a competitive player during the second game.
Simulation:
C
(t−2)+
2∗C
(t−1)3
+cos(2 pi∗rand )∗
√
−2 ln (rand )∗10±5
(2)Where the weighted average of the previous two choices of the participant ( C(t-1) and C(t-2)) is seen as
a mean. A normal distribution is built around this with a Box-Muller transform with a standard deviation of 10, and a random number is chosen from this normal distribution
(
cos(2 pi∗rand )∗
√
−2ln(rand )∗10
). After this 5 is added for the altruistic simulation, whilst 5 is subtracted from the competitive simulation. After rounding to the nearest 10, the action of the simulation is determined. A limit was implemented in which the competitive (altruistic) simulation would not allow the mean to rise (lower) beyond 20 (-20), as well as if both the previous choices were (-) 50 the (competitive) altruistic simulation would simply respond with (-) 50.This allowed for a relatively simple and controlled tit-for-tat simulation of another participant.
TBS procedure
To find the motor threshold, Magnetic stimulation was given over the hand area of the motor cortex using a hand-held figure of eight coil (70 mm standard coil) placed tangentially to the scalp. The handle was placed posteriorly on two subjects and laterally on one subject. Single and repeated pulses were delivered by a Magstim2 stimulator. The stimulation intensity was defined in relation to
the active motor threshold (AMT). This was defined as the minimum single pulse intensity of the motor required to produce a visual twitch on more than five out of ten trials from the contralateral fingers while the subject was maintaining a light voluntary contraction.
The continuous theta burst stimulation(cTBS) protocol of rTMS consisted of bursts containing 3 pulses at 50Hz and an intensity of 80% of the AMT, repeated at 200ms intervals (i.e. at 5Hz). These bursts were administered for 40 seconds (600 pulses).
The pSTS was found by transforming activation areas found in Bault (in preparation) from MNI to the participant’s brain. The coordinates were then adjusted by hand to make sure the stimulation site was accurate, and in grey matter. An NDI Polaris Spectra Neuronavigation system was then used to navigate the coil to stimulate within 2mm of the targeted area. Stimulation was performed with the coil placed tangentially to the skull and the handle pointing posterio-laterally. In the control condition the coil was turned 90 degrees away from the skull so that the lateral edge of the figure of eight coil was still in contact with the skull. The participant was allowed to rest for 10 minutes after the stimulation to allow the cTBS to lower the excitability of the pSTS.
Procedure
Both conditions started with the instructions of the PGBG, including a short test of calculating payoffs to confirm the subjects understood the game, and a test run to acclimatize them with the controls. The subjects also read the instructions of the ring test. The subjects then followed the TBS procedure. After 10 minutes, the subjects performed the ring test of social value orientation while being told that their choices would affect a random participant. The subject then played 20 rounds of the PGBG with the altruistic simulation, followed by a ring test which they were told would affect the person they just played with.
Subsequently, the subject played 20 rounds of the PGBG with the competitive simulation, followed by a ring test which they were told would affect the person they just played with.
The control condition was performed at least 3 days away from the TBS condition, and the order of the conditions was randomized and counterbalanced. One subject performed with the control condition first, while the other two started with the experimental condition first. Due to the low number of participants, the order of the types of simulation could not be counterbalanced. To make sure that the subject believed they were playing against another participant, a Toshiba camera was placed behind them, and turned on before the first ring test. The subject was told that the camera would send a live feed to the other two participants. The participant was asked not to look at the camera, and told that the camera would not film anything on screen. To their right they saw a video recording of two actors, of which the video feed corresponded to the phase of the game they were in: the actors were given explanations of the game, played the game when the participant thought they were playing together, and were idle when the participant was not playing with them. After the final ring test, the camera was turned off, and the subjects were supervised until 90 minutes after the onset of the TBS stimulation for safety purposes.
It was expected that the new PGBG will allow for negative ties to be expressed in the decisions of the participants by taking. Furthermore, decreasing pSTS activity will slow both positive and negative social tie formation due to a hypothesized decrease in the perception of human agency behind the perceived actions. In such we will expect the TBS condition to have, in both the altruistic and competitive simulation, average contributions closer to 0 than their control counterparts, as well as have social ties that are closer to 0 than the controls.
Results:
Although a substantial part of the interactions in the PGBG were takes (21.3%), it is seen that only one instance of taking was not done by the participant with most takes, labeled ME. This participant took during 62.5% of their rounds. These results confirm that the PGBG allows the expression of competitive behaviors and, although there are only three subjects, shows that repeated taking behavior is only seen in a small group of the population.
The average contributions in the negative simulation of the PGBG of subjects WB and KG are higher during the TBS condition than in the sham condition, and equal between conditions for ME. The average contribution in the positive simulation of WB is lower in the TBS condition than in the sham condition, while the average contribution of KG is roughly equal in the TBS condition and the sham condition and higher in the sham condition than in the TBS condition for ME.
ME WB KG
TBS sham TBS sham TBS sham
Positive
simulation -11,5 -16 41,5 48 50 49
Negative
Table 1: Average contributions of the participants in MU
However, when looking at the contributions per round of the participants (see appendix 2-4), the differences between the TBS and sham conditions in WB and KG can be seen to stem from the amount of rounds it takes for the participant to reach a relatively stable contribution, which takes longer during the TBS condition than during the sham condition. However, the longer time to reach a stable contribution coincides with WB and KG’s first sessions. ME shows erratic decisions, switching from positive and negative contributions many times.
The Ring test of social value orientation does not conform to the expectations. We found that for participant ME, the baseline social value orientation angle was positive and similar in both conditions, and near 0 rad for most of the subject specific SVOs. However during her TBS condition and after playing the PGBG with the positive simulation, the SVO angle was -23 degrees, indicating that they would prefer to lose earnings to lower the earnings of the positive simulation. However, the length of the vector of the SVO (515MU) was under the threshold of 600MU that Liebrand (1984) has set, indicating that the responses given were too erratic to include.
ME WB KG
TBS sham TBS sham TBS sham
SVO 6,87 8,25 0 0 3,67 0
pos -23,15 -2,86 3,61 0 -3,67 0
Neg 0,06 3,78 0 0 0,26 0
Table 2: Social value orientation in degrees
The ring tests for WB and KG show low SVO angles, which do not conform to the expectations seen in previous experiments: SVO values seem not to change after positive and negative economic
interactions, or even decrease after positive economic interactions.
The average reaction times of the PGBG (see table 3) seem not to differ between TBS and sham conditions (t(5) = -,85, p=0,43). The largest differences can be seen with WB’s and ME’s positive sessions, in which the longer RT’s are also the subjects’ first sessions.
The average reaction times in the ring test (see table 4) are longer in the TBS than the sham condition of WB and KG, while longer in the sham than TBS session of ME. These longer RT’s also coincide with the participants’ first sessions.
TBS positive TBS negative sham positive sham negative WB 2914 2504 2750 2567 KG 2765 2889 2740 2922 ME 2724 2658 3187 2724
TBS Sham
ME 1286 2319
WB 3005 2651
KG 2598 1865
Table 4: reaction time of the ring test in ms
WB ME KG Empathic concern 2,29 1,71 2 Fantasy scale 2,29 1,29 2,57 Personal distress 1,71 1,71 1,57 Perspective taking 2,14 2,14 2,57
Table 5: Interpersonal reactivity scale (0-4)
WB ME KG
Social dominance 3,56 2,31 2,63
Table 6: Social dominance orientation from 1(not dominant) - 7 (very dominant)
Qualitatively, the only link found between the interpersonal reactivity scales (table 6) and the contributions during the PGBG lie between a low fantasy scale and a lower contribution.
All subjects have stated their doubts about actually playing against another human during the task, while reasoning that all previous social psychological experiments they participated in included some form of deception about the person they were doing the experiment with.
Discussion:
The results support that the PGBG allows participants to demonstrate both pro-social and competitive behavior. The results also support the previous finding that only a distinct amount of people would show competitive behavior, whilst taking does occur during different strategies.
The participants were able to drive their own behavior according to their own will, whilst still being affected by the other player’s decisions. However, while the participants found the simulation of the other person believable, it is no substitute for playing with a real human being.
The main effects of the TBS point towards there needing to be more rounds of the game before the participants arrive to a stable contribution choice. This could be explained by the pSTS being an integral part of updating the social tie between the participant and their opponent, in which the impulse resulting from their opponent’s previous choice becomes diminished, or that their tie persistence is increased, both of which would result in a slowed update of the social tie. However, this longer time could also be explained with an order effect. Although the experiment was counterbalanced to tease out any order effects, due to the small number of subjects and that both participants with pro-social tendencies were in the TBS condition first, the increased time to arrive to a stable contribution choice could be explained by a simple learning effect.
It is always difficult to validate a TBS experiment: its main effects on cognition are usually subtle, and are difficult to see with such low statistical power. The side-effect of TBS, and most forms of offline TMS, however can be seen as a proxy as to the TBS being effective. This lies within an increased RT. In this experiment we have seen an increase in response time during the ring test for the subjects KG and WB, whilst finding a largely decrease RT for ME. However, similar to the alternative explanations for the increased time for coming to a stable contribution choice, KG and WB were in the randomized group that was subjected to the TBS condition first, bringing in the possibility of training effects which could account for the increased RT. Usually this would be accounted for by the counterbalance, however, due to ME’s outlying result, and the low power, nothing can be concluded from the ring test RTs.
The more erratic SVO results of the ring test in the TBS condition compared to the control condition however, were not expected. It is of yet unclear if this is due to a discrepancy between the
hypothesized effect of TBS of the pSTS on the development of social ties during the PGBG, or due to one of the side effects of TBS.
However, this does cast a shadow of doubt upon the use of the ring test in conjunction with TBS: in future experiments it might be better to use the data of the participants, and estimate the social tie using the economic model.
The RT of the PGBG was designed so that the participants are forced to take 2000ms after being presented with the choice screen to decide what they will do in that particular round. Although this method ensures that the participants will think about their decision, it also raises the lower bound of the RT, working as a confounding factor for recording the correct results. Perhaps it would be better to remove this constraint whilst still instructing the participant to think about their choice.
Whilst there was a camera trained upon the participants, and the participants were shown videos of actors as their opponents, most participants expressed their disbelief in playing against an actual other participant due to previous psychology experiments in which they had participated. While they did report the feeling of playing against a real player during the game, they still held the belief that it was a simulation. An added effect that might have been introduced by the camera is the feeling of being watched. This could lead to a skewing towards pro-social behavior (Bateson 2006)
A possible improvement could be seen by placing the 2 participants, or a participant and an actor in the same room when the experiment is performed. To make sure other biases from when the other participant is seen and sees the other player do not interfere, the experiment could be done in a group setting of 4.
The PGBG shows potential in allowing participants to allow competitive, pro-social, and selfish behaviors come out and build upon itself during an interaction with a real person. An inherent con in this experiment lies the dependency upon many subjects, due to both the many different pairings of types of behaviors and the many smaller factors in the beginning of an interaction that can have a large effect on the results.
In all however, the results of the PGBG under TBS does weakly point towards our prediction of TBS: decreasing contributions against a pro-social opponent; and increasing contributions against a competitive opponent. It points towards the pSTS being involved in social decision making and social ties. While due to insufficient data in particular conditions, there is no way to separate a possibility of a TBS effect from training effects, primacy effects and spillover effects in this experiment, it seems to be a good design in which to conduct future research.
Reference List:
Abbink K, Sadrieh A. (2009) The pleasure of being nasty. Economics Letters; 105: 306-8.
Allison, T., Puce, A., McCarthy, G. (2000), "Social perception from visual cues: role of the STS region", Trends in Cognitive Sciences, Vol. 4, No. 7, pp. 267-278.
Bateson, M., Nettle, D., Roberts, G. (2006) Cues of being watched enhance cooperation in a real-world setting, Biology Letters, 2:412-414
Batson, C. D. (1991). The altruism question: towards asocial-psychological answer. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Bault N, Pelloux B, Fahrenfort J, Ridderinkhof KR, van Winden FAAM. Neural underpinnings of dynamic social ties formation in a public good game. In preparation.
Bault N, Coricelli G, Rustichini A. Interdependent utilities: How social ranking affects choice behavior. PLoS ONE 2008; 3: e3477.
Behrens, T.E.J., Hunt, L.T., Rushworth, M.F.S.(2009) The computation of social behavior. Science, Vol 324, No. 5931, pp. 1160-1164.
Berg, J., Dickhout, J., & McCabe, K. (1995). Trust, Reciprocity and Social History. Games and Economic Behaviour, 10, 122-142.
Bertini, C., Leo, F., Avenanti, A., Làdavas, E. (2010). Independent mechanisms for ventriloquism and multisensory integration as revealed by theta-burst stimulation. European Journal of Neuroscience, Vol. 31, pp. 1791–1799, 2010
Bosman R, Van Winden F. EMOTIONAL HAZARD IN A POWER-TO-TAKE EXPERIMENT. The Economic Journal 2002; 112: 147-69.
Charness G, Masclet D, Villeval M. (2010) Competitive preferences and status as an incentive: Experimental evidence. SSRN eLibrary .
de Bruijn ERA, de Lange FP, von Cramon DY, Ullsperger M. (2009). When errors are rewarding. Journal of Neuroscience, 29: 12183-6.
De Dreu, C.K.W. (2012) Oxytocin modulates cooperation within and competition between groups: an integrative review and research agenda. Hormones and behavior. Vol 61(3), 419-429.
de Quervain, D.J.F., Fischbacher, U., Treyer, V., Schelthammer, M., Schnyder, U., Buck, A., & Fehr, E. (2004). The Neural Basis of altruistic punishment. Science, 305(5688), 1254-1258.
Decety J, Jackson PL, Sommerville JA, Chaminade T, Meltzoff AN. (2004) The neural bases of cooperation and competition: An fMRI investigation. NeuroImage, 23: 744-51.
Di Pellegrino G., Fadiga L., Fogassi L., Gallese V., Rizzolatti G. (1992).Understanding motor events: A neurophysiological study. Experimental Brain Research, 91, 176–180.
Fahrenfort, J.J., Van Winden, F., Pelloux, B., Stallen, M., and Ridderinkhof, K.R. (2012) Neural correlates ofdynamically evolving interpersonal ties predict prosocial behavior. Frontiers in Neuroscience, vol 6, 1-14.
Fehr, E. & Gachter, S. (2000). Cooperation and punishment in public goods experiments. American Economic Review, 90(4), 980-994.
Frith, C.D. & Frith, U. (2006). How we predict what other people are going to do. Brain research 1079: 36-46.
Gallagher, H.L., Jack, A.I., Roepstorff, A., Frith C.D. (2002) Imaging the intentional stance in a competitive game NeuroImage, 16 (3): 814–821
Gazzola V., Aziz-Zadeh L., Keysers C. (2006). Empathy and the somatotopic auditory mirror system in humans. Current Biology, 16, 1824–1829.
Hampton, A.N., Bossaerts, P., and O’doherty, J.P. (2008). Neural correlates of mentalizing-related computations during strategic interactions in humans. Proc.Natl.Acad. Sci. U.S.A. 105, 6741–6746.
Harbring C, Irlenbusch B, Kräkel M, Selten R. (2007) Sabotage in corporate contests : An experimental analysis. International Journal of the Economics of Business. 14: 367-92.
Haruno, M., and Kawato, M. (2009). Activity in the superior temporal sulcus highlights learning competence in an interaction game. J. Neurosci. 29, 4542–4547.
Krueger, F., McGabe, K., Moll, J., Kriegeskorte, N., Zahn, R., Strenziok, M., Heinecke, A., & Grafman, J. (2007). Neural Correlates of trust. Proceedings of the National Academy of Sciences of the USA, 104 (50), 20084-20089
Lamm, C., Batson, C. D. and Decety, J. (2007), "The neural substrate of human empathy:
effects of perspective-taking and cognitive appraisal", Journal of Cognitive Neuroscience, Vol. 19, No. 1, pp. 42-58.
J. Ledyard Public goods: a survey of experimental research J. Kagel, A. Toth (Eds.), Handbook of Experimental Economics, Princeton University Press, Princeton, NJ (1995)
Liebrand, W. B. G. (1984). "'The effect of social motives, communication and group size on behaviour in an n-person multi stage mixed motive game". European Journal of Social Psychology 14 (3): 239– 264.
McCabe, K., Houser, D. Ryan, L., Smith, V., Trouard, T. (2001) A functional imaging study of
cooperation in two-person reciprocal exchange Proc. Natl. Acad. Sci. U. S. A., 98 (20) 11832–11835 Moll, J., Krueger, F., Zahn, R,. Pardini, M., de Oliviera-Souza, R., & Grafman, J. (2006). Human fronto-mesolimbic networks guide decisions about charitable donation. PNAS 103 (42): 15623-15628. Ramnani N, Miall RC: A system in the human brain for predicting the actions of others. Nat Neurosci 2004, 7:85-90.
Rilling, J. K., Sanfey, A. G., Aronson, J. A., Nystrom, L. E. and Cohen, J. D. (2004), "The neural correlates of theory of mind within interpersonal interactions", Neuroimage, Vol. 22, No. 4, pp. 1694-703.
Saxe, R. Uniquely human social cognition. Current Opinion in Neurobiology 2006, 16:235–239 Singer, T.,Kiebel, S.J., Winston, J.S., Dolan, R.J., and Frith, C.D.(2004). Brain responses to the acquired moral status of faces. Neuron 41, 653–662.
Tankersley D, Stowe CJ, Huettel SA. (2007) Altruism is associated with an increased neural response to agency. Nature Neuroscience 10: 150-1.
van Dijk,F.,Sonnemans,J.,andvan Winden,F.(2002).Socialtiesina public goodexperiment. J. Public Econ. 85, 275–299.
Van Kemenade, B.M., Muggleton, N., Walsh, V., Saygin, A.P. (2012). Effects of TMS over Premotor and Superior Temporal Cortices on Biological Motion Perception. Journal of Cognitive Neuroscience 2012 24:4, pp. 896–904
Zeki S, Romaya JP (2008) Neural Correlates of Hate. PLoS ONE 3(10): e3556. doi:10.1371/journal.pone.0003556
Zilbovicius et al. (2006) Autism, the superior temporal sulcus and social perception
Zizzo DJ. The cognitive and behavioral economics of envy. In: Smith RH, editor. Envy: Theory and research. Oxford: Oxford University Press; 2008. p. 190-210.
Zizzo DJ. Money burning and rank egalitarianism with random dictators. Economics Letters 2003; 81: 263-6.
Zizzo DJ, Oswald AJ. Are people willing to pay to reduce others' incomes? Annales d'Économie et de Statistique 2001; 63/64: 39-65.
Appendices:
Appendix 1: Payoff matrix PGBG
Appendix 2: Results PGBG KG
Appendix 3: Results PGBG WB
Appendix 4: results PGBG ME
