Rogerslab.org.uk

Neuropsychopharmacology (2006) 31, 1075–1084& 2006 Nature Publishing Group Effects of Tryptophan Depletion on the Performance of anIterated Prisoner’s Dilemma Game in Healthy Adults Richard M Wood1, James K Rilling2, Alan G Sanfey2, Zubin Bhagwagar3 and Robert D Rogers*,1 University Department of Psychiatry, University of Oxford, Oxford, UK; 2Departments of Anthropology, Psychiatry and Behavioral Sciences, Emory University, Atlanta, GA, USA; 3Department of Psychiatry, Yale University, New Haven, CT, USA Adaptive social behavior often necessitates choosing to cooperate with others for long-term gains at the expense of noncooperative behaviors giving larger immediate gains. Although little is know about the neural substrates that support cooperative over noncooperative behaviors, recent research has shown that mutually cooperative behavior in the context of a mixed-motive game, the Prisoner’s Dilemma (PD), is associated with increased neural activity within reinforcement circuitry. Other research attests to a role for serotonin in the modulation of social behavior and in reward processing. In this study, we used a within-subject, crossover, double-blind design to investigate performance of an iterated, sequential PD game for monetary reward by healthy human adult participants following ingestion of an amino-acid drink that either did (T + ) or did not (TÀ) contain l-tryptophan. Tryptophan depletion produced significant reductions in the level of cooperation shown by participants when playing the game on the first, but not the second, study days. This effect was accompanied by a significantly diminished probability of cooperative responding given previous mutually cooperative behavior.
These data suggest that serotonin plays a significant role in the acquisition of socially cooperative behavior in human adult participants, and suggest novel hypotheses concerning the serotonergic modulation of reward information in socially cooperative behavior in both Neuropsychopharmacology (2006) 31, 1075–1084. doi:10.1038/sj.npp.1300932; published online 11 January 2006 Keywords: Prisoner’s Dilemma; serotonin; reciprocal cooperation; social function; aggression; reward evolutionary contexts (Axelrod, 1984; Axelrod and Hamil-ton, 1981), individual differences in competitive and Mutually cooperative behavior among individuals is an exploitative strategies during reciprocal interactions (Pruitt integral part of greater-ape society, yet there has been and Kimmel, 1977), and to identify factors that promote or relatively limited research into the nature of the neural inhibit cooperation in diverse experimental and naturalistic substrates supporting social cooperation in human and settings (Axelrod, 1984; Boone et al, 2002; Sheldon, 1999).
non-human species. The Prisoner’s Dilemma (PD) offers a Rilling et al (2002) used functional magnetic resonance tractable laboratory measure of social cooperation based imaging (fMRI) to investigate regional changes in blood- upon reciprocal altruism which can be used to advance this oxygenation-level-dependent response (BOLD) associated area experimentally (Trivers, 1971, 1985): two players make with socially cooperative behavior in an iterated PD game choices associated with certain pay-offs (represented by a (viz., a game which is played several consecutive times or ‘pay-off matrix’; see below). The pattern of choices made ‘rounds’ with the same partner). Mutually cooperative can lead to an equal sharing or an unequal sharing of the outcomes (in which both players chose to cooperate: CC) pay-offs (favoring one player or the other), thereby allowing were associated with increased BOLD within the orbito- players to display cooperation or selfishness. This model frontal cortex compared to mixed outcomes (in which one has been used extensively by researchers from many of the two players chose to cooperate, while the other chose disciplines to explore the emergence of cooperation in to defect: CD or DC) and mutual defection outcomes (inwhich both players chose to defect: DD). Although this was *Correspondence: Dr RD Rogers, Department of Psychiatry, Uni- the case when players believed they were playing with a versity of Oxford, Warneford Hospital, Oxford, Oxfordshire OX3 7JX, human partner or a computer, mutual cooperation was UK, Tel: + 44 1865 226 399, Fax: + 44 1865 793 101, associated with greater activation in the rostral anterior cingulate cortex and anteroventral striatum specifically Received 23 December 2004; revised 21 July 2005; accepted 22August 2005 when participants believed they were playing with a human Online publication: 9 September 2005 at http://www.acnp.org/citations/ partner. These neural systems receive dopamine projec- tions from the midbrain and are known to be involved Effects of tryptophan depletion on PD game performance in reward and the processing of motivationally significant Chronic citalopram was associated with increased choices information (Robbins and Everitt, 1996), suggesting that of the more generous pay-offs and increased affiliative socially cooperative behavior is supported by activity within messages sent to the playing partner (Tse and Bond, 2002a).
reinforcement pathways (Moskowitz and Cote, 1995). A In a similar study, reboxetine increased cooperation in the follow-up study using a one-shot PD paradigm in which PD game, while citalopram sustained emotional expression participants played a new partner on each round con- of speech when speaking to a socially withdrawn partner firmed this hypothesis by demonstrating increased BOLD (Tse and Bond, 2002b). Summarizing the above results, within subcallosal anterior cingulate cortex, anteroventral increasing serotonin activity appears to change indirect striatum, and medial orbitofrontal cortex with mutually measures of social function in healthy human adults, and cooperative outcomes (Rilling et al, 2004).
may increase affiliative gestures in dyadic interactions.
Observations in several species indicate a role for However, there is little information about the effects of serotonin in the modulation of prosocial behavior. Inves- reducing serotonin on cooperative behavior in mixed- tigation of both peripheral and central indices of serotonin motive games, such as a true iterated PD, where the gains function have shown negative associations with violent and and losses for one player depend upon effective reciprocal aggressive behavior in rhesus monkeys (Higley et al, 1992, exchanges with a social partner. We sought to investigate 1996) and human clinical samples (Brown et al, 1979; Virkkunen et al, 1994), as well as positive associations with We used rapid tryptophan depletion to restrict the socially affiliative behaviors (such as grooming and availability of l-tryptophan for serotonin synthesis in approach) (Raleigh et al, 1981). Pharmacological interven- healthy human participants in order to explore subsequent tions that increase serotonergic activity also attest to a changes on performance of an iterated PD game for prominent role in important aspects of social behavior and monetary reward. This methodology has shown to be an aggression (for a discussion, see Fuller, 1996). The selective effective way of temporarily reducing serotonin activity in serotonin reuptake inhibitor (SSRI), fluoxetine, increases both animal and humans (Biggio et al, 1974; Carpenter et al, affiliative behaviors in rodents (Knutson and Panksepp, 1998), with evidence of cognitive effects in healthy 1996), while each of fluoxetine, quipazine (a 5-HT1 and participants without a history of mood disorder (Park 5-HT2 receptor agonist), and the amino-acid serotonin et al, 1994; Rogers et al, 1999; Moore et al, 2000). The precursor, l-tryptophan, have been found to increase starting point for our study was the accumulating evidence affiliative behaviors and decrease nonsocial behaviors (such that serotonin plays a role in modulating the various aspects as vigilance and avoidance) in primates (Raleigh et al, of reward processing (Aronson et al, 1995; Redgrave and 1985). Additional information about these effects has been Horrell, 1976; Sasaki-Adams and Kelley, 2001) mediated by provided by observations that fluoxetine and l-tryptophan circuitry encompassing orbitofrontal and rostral cingulate increase the social dominance of subdominant vervet males cortices, and ventral striatum (Robbins et al, 1989). We in the absence of the alpha male, while interventions that wished to test the hypothesis that reducing serotonergic antagonise serotonin (eg the 5-HT2C receptor antagonist, modulation in healthy participants would reduce the cyproheptadine, and chronic treatment with fenfluramine) number of cooperations and increase the number of prevent the emergence of dominance under these condi- defections in an iterated PD game, perhaps by interfering with activity within reinforcement-related circuitry asso- In humans, major depressive disorder is widely accepted ciated with mutually cooperative behavior.
as involving altered serotonin function and is characterizedby relatively poor social functioning (Fossi et al, 1984;Weissman, 2000). Consistent with this finding, antidepres- sant medications that enhance serotonergic function havebeen shown to increase affiliative behaviors, including The study was approved by the Oxfordshire Psychiatry verbal and nonverbal positive communication in depressed Research Ethics Committee, UK. All participants gave patients, sometimes independently of effects on depressed mood (Dubini and Polin, 1997), and can reduce hostilesentiment and violent outbursts in impulsive clinical samples (Coccaro and Kavoussi, 1997).
Despite this, experimental investigation of the role of A total of 24 healthy adults (12 males; 12 females) serotonin in the social behavior of human participants has participated. Previous research has shown that tryptophan been limited. Chronic l-tryptophan intake has been found to depletion in participants with a history of mood disorder both increase dominant behaviors, and decrease quarrel- can reinstate dysphoria (Moore et al, 2000). Therefore, someness, in everyday interactions (Moskowitz et al, 2001), participants were carefully examined by an experienced while the SSRI, paroxetine, has been found to increase psychiatrist (ZB) to ensure that none of the following affiliative gestures towards a partner during completion of a exclusion criteria were met: (i) major physical illness, (ii) puzzle task after 1 but not 4 weeks of treatment (Knutson current or previous DSM-IV major mood disorder as et al, 1998). More formally, Tse and Bond (2002a, b) have assessed by a SCID-I interview (First et al, 2002), and (iii) examined the effects of chronic treatment with the SSRI, current or previous substance abuse. Participants with a citalopram, and the selective noradrenergic reuptake family history of mood disorder were not excluded. There inhibitor, reboxetine, on the performance of an amended was no restriction on the phase of menstrual cycle in female PD game involving the division of pay-offs between the two participants. Verbal IQ was estimated with the National players and communication with the playing partner.
Effects of tryptophan depletion on PD game performanceRM Wood et al game, each participant was introduced to a gender-matched‘partner’Fin reality, an experimenter-confederate. It was The study consisted of a within-subject, crossover, double- explained to the participant that his/her computer was blind design. Each participant ingested an amino-acid drink networked to a computer in another room in which their on 2 study days, separated by not less than 1 week. One partner would play. The experimenter left to take the drink contained l-tryptophan (T + ), the other drink did not confederate to their room, returning a few minutes later to (TÀ). The order of these treatments was counterbalanced talk the participant through the PD game instructions.
across the participant group. A total of 12 participants under- In fact, the confederate’s choices were made by the went the T + treatment first and the TÀ treatment second computer, running a predetermined strategy of tit-for-tat (six male and six female) (T + /TÀ). A total of 12 participants underwent these treatments in the reverse order (six male On each round of the PD game, the participant viewed a 2 Â 2 matrix that displayed the pay-off matrix for that round(see Figure 1). The four squares of this pay-off matrix defined four possible outcomes: both players cooperate(CC); the participant cooperates but the partner defects Amino acids were supplied by SHS International Ltd.
(CD), the participant defects but the partner cooperates The amounts of each for male and female participants, (DC), and both players defect (DD). The payoffs for the respectively, were l-alanine (5.5 g; 4.58 g), l-arganine (4.9 g; outcomes were constructed such that DC4CC4DD4CD, 4.08 g), l-cystine (2.7 g; 2.25 g), glycine (3.2 g; 2.67 g), and CC4(CD + DC)/2 (2). Specifically, if the participant l-isoleucine (8.0 g; 6.67 g), l-leucine (13.5 g; 11.25 g), l-lysine and partner cooperated (CC), both would receive 20p. If monohydrochloride (11.0 g; 9.17 g), l-methionine (3.0 g; the participant defected while the partner cooperated (DC), 2.5 g), histidine (3.2 g; 2.67 g); l-phenylalinine (5.7 g; 4.75 g), the participant would earn 30p, and the partner nothing; l-proline (12.2 g; 10.17 g), l-serine (6.9 g; 5.75 g), l-threonine and vice versa (CD). If both players defected (DD), both (6.5 g; 5.42 g), l-tyrosine (6.9 g; 5.75 g), l-valine (8.9 g; 7.42 g).
The T + drink contained l-tryptophan (2.3 g; 1.92 g). The The participant’s partner always chose first (ie the game unpleasant taste of the drinks was masked using 5 g flavor was sequential). If the partner cooperated, the upper row of sachets. Each sachet consisted of citric acid (or malic acid), the matrix was highlighted in yellow. If the partner defected, flavorings (lemon-and-lime, cherry-and-vanilla, or grape the lower row was highlighted in yellow. After this, the fruit), and artificial sweetener. Each sachet contained participant responded with his own choice. If he/she approximately 15 calories, and 1.3 g of carbohydrate.
cooperated, the left column was highlighted; if he/shedefected, the right column was highlighted. Consequently, the combination of the partner’s and the participant’schoices specified the outcome of the round. The participant Participants followed a low-protein diet (less than 20 g) the pressed the ‘C’ key on the computer keyboard to indicate a day before each study day. Participants attended the cooperative response, and the ‘D’ to indicate a defection laboratory at 0830 hours. Blood samples were taken to response. Before starting, four training trials demonstrated obtain baseline levels of plasma total l-tryptophan. Partici- the four possible choice combinations; the participant was pants then drank the amino-acid drink over a 30-min also shown a table showing the total earnings arising period. Some participants reported transitory nausea through the four possible choice outcomes over the course and tiredness. Participants were given a low-protein lunch (less than 2 g). At 5 h after consuming the amino-acid drink( + 5 h), a second blood sample was taken. Participantsthen completed the PD game.
Participants completed self-report measures of state positive and negative affect (PANAS; Watson et al, 1988)and validated visual analog scales of aggression (VAS; Bondand Lader, 1974) at baseline and + 5 h. Participants alsocompleted psychometric assessments of trait aggressionusing the Buss–Perry Aggression Questionnaire (AQ; Bussand Perry, 1992). The AQ provides subscales for physicaland verbal aggression, hostility, and anger. The VAS wasused to assess effects of tryptophan depletion on partici-pants’ mood, while the AQ was used to assess whethereffects on social cooperation involved differences in traitaggression.
The iterated, sequential PD game. The iterated, sequentialPD used in the study was adapted from that reported by Payoff matrix for the four outcomes in the iterated, sequential Prisoner’s Dilemma (PD) game. The participant’s choices (‘You’) are listed Rilling et al (2002). The game was presented on a computer atop columns and the partner’s choices (‘partner’; in fact, a computer that had connecting cables going into the laboratory wall, program playing tit-for-tat) are listed aside the rows. The payoffs for each facilitating the deception that the computer was networked player, depending upon both players’ choices, are shown within each to a different computer in another room. Before playing the square; green ¼ participant’s and pink ¼ partner’s).
Effects of tryptophan depletion on PD game performance Table 1 Monetary Payoffs for Four Different Strategies Over the tion of the data from the PD games.) Where the effects of Course of the Iterated, Sequential Prisoner’s Dilemma (PD) Game the TÀ and T + treatments were different on the first and second study days, as indicated by significant two-wayinteractions between treatment and treatment order, dependent measures were subject to supplementary ANO-VAs for each study day separately (to investigate simple Both you and your partner cooperate each round interaction effects), with gender and treatment (T + vs TÀ) Your partner cooperates and you defect each round as two between-subject factors and game as a single within- Your partner defects and you cooperate each round Both you and your partner defect each round ‘You’ ¼ participant; ‘Partner’ ¼ computer program playing tit-for-tat.
Participants’ mean age and verbal IQ were 27.0471.49 (SE) Participants played two separate PD games, each consist- and 113.8771.43. There were no significant differences ing of 20 rounds. The procedure contained two manipula- between the age or estimated IQs of those participants who tions. First, the computer started the first game by choosing completed the T + /TÀ treatment order (27.8972.26 years to cooperate but started the second game by choosing to and 11271.98) and those who completed the TÀ/T + order defect. Second, after its first choice in both games, the (26.2072.00 and 11671.98) (Fs(1, 20)o2.35). Similarly, computer followed a strict tit-for-tat strategy, always the two groups were matched for trait positive affect mimicking the responses of the participants on the previous (36.8371.52 vs 33.9271.26; F(1, 20) ¼ 2.45), trait negative round. Participants were thoroughly debriefed about the affect (15.3371.80 vs 13.2570.70; F(1, 20) ¼ 1.14), total self- deception at the end of the study; all indicated that they had report aggression (6272.45 vs 66.1874.14; Fo1.00), believed that they had been playing the PD game with a real physical aggression (14.2570.90 vs 17.0971.88; F(1, 15) ¼ human partner (ie the confederate) rather than a computer 1.31), verbal aggression (15.0070.46 vs 16.0970.63; F(1, 15) ¼ 1.61), anger (14.3870.98 vs 16.3671.27; F(1, 15) ¼1.15), and hostility (18.3871.21 vs 16.6471.30; F(1, 15) ¼1.04). (For technical reasons, scores from the AQ (Buss and Perry, 1992) were unavailable for four participants who The dependent measures for the PD game were as follows: completed the T + /TÀ treatment order and for one (i) the proportion of rounds on which participants chose participant who completed the TÀ/T + treatment order.) to cooperate (‘proportionate choice’); (ii) the deliberationtime (in milliseconds) taken to make a cooperative or a defection choice; (iii) the ‘conditional probability’ ofcooperating specified as the proportion of rounds on which As expected (Moore et al, 2000), total plasma tryptophan participants cooperated following each of the four possible was reduced between baseline and + 5 h after the TÀ outcomes (CC, CD, DC, DD) on the immediately previous treatment (12.4470.68 vs 1.9270.27 mg/ml) but increased after the T + treatment (13.0170.51 vs 30.2772.31 mg/ml) Differences in age, estimated verbal IQ, and psychometric (F(1, 12) ¼ 123.43, po0.0001). Analysis of simple effects measures of self-report trait affect and aggression were confirmed that the difference in plasma tryptophan follow- analyzed by analysis of variance (ANOVA) with treatment ing the TÀ and T + treatments was significant at + 5 h order across the two study days (T + /TÀ and TÀ/T + ) and Proportionate choice of cooperative responses were analyzed with repeated measures ANOVAs with gender,treatment order as between-subject factors, and treatment Treatment with the TÀ drink produced different kinds of (T + vs TÀ) and game (1 vs 2) as within-subject factors. The effects on cooperative behavior on the 2 study days as reaction times (ms) for cooperative and defection responses evidenced by a significant two-way interaction between were averaged together to give mean deliberation times and treatment and treatment order (F(1, 20) ¼ 4.68, po0.05).
submitted to the same analysis as the proportionate choice Statistical analysis of each study day separately demon- data. Plasma tryptophan, state positive affect, state negative strated that the TÀ treatment was associated with affect (PANAS), and state aggression (VAS) were analyzed significantly fewer cooperative responses than the T + with repeated measures ANOVAs with gender and treat- treatment on day 1 (F(1, 20) ¼ 6.67, po0.05) (see Figure 2), ment order as between subject-factors and treatment (T + but a slightly increased number of such responses on day 2 vs TÀ) and time (baseline vs + 5 h) as within-subject (F(1, 20) ¼ 2.63) (Figure 2). Pair-wise tests showed that factors. (The 13 VAS subscales were included in order to participants who underwent the TÀ treatment on day 1 assess whether the performance of the PD games following showed a significant increase in cooperative responses after tryptophan depletion might be attributable to short-term the T + treatment on day 2 (0.5070.07 to 0.6470.08; changes in aggression. We did not correct for multiple t (two-tailed) ¼ 2.26, df ¼ 11, po0.05), while participants comparisons when analyzing this data in order to highlight who underwent the T + treatment on day 1 showed any such changes that might be relevant to the interpreta- no significant change in cooperative responses after the Effects of tryptophan depletion on PD game performanceRM Wood et al TÀ treatment on day 2 (0.7370.05 to 0.8070.04; t (two- However, analysis of each study day separately showed that tailed) ¼ À0.98, df ¼ 11, po0.35).
the deliberation times following the TÀ and T + treatments Overall, there was a trend for all participants to make did not differ significantly on day 1 (32497421 ms vs more cooperative responses on the first game of 20 rounds 29957387 ms, respectively) (Fo1.00) or day 2 (23377 than the second game of 20 rounds (0.7470.05 vs 279 ms vs 20427208 ms; Fo1.00). Mean deliberation times 0.5970.06), (F(1, 20) ¼ 3.49, p ¼ 0.08). There were no in the first game of 20 rounds were not significantly significant differences between male and female participants different from those in the second game (28607268 vs in the proportion of cooperative responses (0.6870.06 vs 24527190 ms) (F(1, 20) ¼ 2.94). Male participants did not 0.6570.06, respectively; Fo1.0). There were no significant significantly differ from female participants in their interactions involving treatment, gender, and game.
deliberation times (27567359 vs 25577197 ms) (Fo1.00).
Treatment with the TÀ drink affected participants’ There were no other significant interactions involving deliberation times differently on the first and second study days, as evidenced by a two-way interaction between Conditional probabilities of cooperative responses were treatment and treatment order (F(1, 20) ¼ 15.42, po0.005).
defined as the proportion of trials on which participantsmade a cooperative choice given one of the four outcomes(CC, CD, DC, and DD) on the previous round of the game.
On the first study day, the TÀ participants exhibited asignificantly reduced conditional probability of a coopera-tive response given a mutual cooperative outcome on theprevious round compared to the T + treatment (seeFigure 3; F(1, 22) ¼ 4.35, po0.05); no other differencesin the conditional probability of cooperative responseswere significant (all Fsp1.07). On the second day, theTÀ treatment produced a slightly increased conditionalprobability of a cooperative response given a mutual co-operative outcome on the previous round compared to theT + treatment; this effect was not significant (Figure 3;F(1, 22) ¼ 2.42).
Overall, state positive affect was significantly reducedbetween baseline and + 5 h (F(1, 20) ¼ 15.88, po0.005),but no more so after the TÀ treatment (31.2971.42 vs Mean proportion of cooperative choices of healthy participants 28.9671.47) compared to the T + treatment (30.4271.61 vs who drank an amino-acid drink containing l-tryptophan (T + ) treatment and participants who drank an amino-acid drink not containing l-tryptophan 71.54) (Fo1.00). State negative affect did not change (TÀ treatment) on the first and second study days. * ¼ F(1, 20) ¼ 6.67, markedly between baseline and + 5 h (F(1, 20) ¼ 2.29) after either the TÀ treatment (12.7170.94 vs 12.4671.01) or T + The conditional probability of making a cooperative choice given the four possible outcomes of the previous round in participants who drank an amino-acid drink containing l-tryptophan (T+) and in participants who drank an amino-acid drink not containing l-tryptophan (TÀ) on the first and secondstudy days (CC ¼ participant cooperates–partner cooperates, CD ¼ participant cooperates–partner defects, DC ¼ participant defects–partner cooperates,DD ¼ participant defects–partner defects). * ¼ F(1, 21) ¼ 4.352, po0.05.
Effects of tryptophan depletion on PD game performance treatment (13.5471.02 vs 12.6370.69) (Fso1.00). There TÀ compared to the T + treatment on either study day (see were no significant differences in the state positive or Table 2). Moreover, the reduced cooperation of those negative affect of the T + and TÀ participants at baseline participants tested after the TÀ treatment on study day 1 or + 5 h on either study day 1 or day 2 (all Fso1.00; cannot be attributed to an increase in state aggression in see Table 2). Entering state positive and negative affect comparison with those tested after the T + treatment. These at + 5 h on day 1 as covariates did not abolish the two groups of participants were matched on the multiple significant reduction in cooperative responses associated facets of aggression measured by the VAS (Bond and Lader, with the TÀ compared to T + treatments (F(1, 18) ¼ 6.70, 1974) at baseline, and showed broadly similar changes during the interval between baseline and playing the PDgame + 5 h later. At that time, 11 of the 13 VAS showedlower scores associated with the TÀ treatment compared to the T + treatment and only two showed higher scores (see Overall, self-reported annoyance was significantly decreased Table 2). Indeed, the single significant treatment difference after the TÀ compared to the T + treatment (21.7772.62 vs (uncorrected for multiple comparisons) at the time of 26.0473.49) (F(1, 20) ¼ 4.88, po0.05); self-reported disgust playing the PD game on day 1 indicated that participants was also reduced after the TÀ treatment compared to the tested after the TÀ treatment rated themselves as signifi- T + treatment (21.8172.74 vs 29.1073.64) (F(1, 20) ¼ 5.48, cantly less rebellious than those tested after the T + po0.05). Comparisons of ratings at + 5 h on study day 1, treatment. In summary, there is no evidence to suggest taken just before playing the PD game, revealed that that the reduced cooperation associated with the TÀ treat- rebelliousness was significantly lower following the TÀ ment on the first study day is explicable by differences in state affect, or increased state aggressivity, consequent state furiousness showed a similar trend (F(1, 20) ¼ 3.48, to tryptophan depletion. Rather, the data indicate that p ¼ 0.08). There were no other significant treatment-related serotonin can affect social cooperative behavior indepen- differences in state aggression ratings at baseline or + 5 h dently of the gross changes in emotional state that might be detectable by self-report measures of positive or negativeaffect including state aggression.
Similarly, those participants tested after the TÀ treatment and those tested after the T + treatment on the first studyday also scored equivalently on trait positive and negative The iterated PD has been used extensively by researchers in affect, as well as on self-report trait physical aggression, social psychology, economics, and game theory to model trait verbal aggression, trait anger, and trait hostility.
social relationships involving reciprocal altruism (Axelrod, Consequently, the reduced cooperation in the PD game 1984; Axelrod and Hamilton, 1981; Trivers, 1971, 1985).
associated with the TÀ treatment cannot be the result of These results demonstrate that healthy adult participants between-subject variation in personality traits reflecting who ingested an amino-acid drink lacking l-tryptophan stable aspects of emotional function and aggression. (The (TÀ treatment) showed a pattern of reduced cooperative two participant groups were also closely matched for age responses on an iterated, sequential PD game compared and estimated verbal IQ.) Accumulating data suggest that to participants who drank a tryptophan-balanced drink the effects of tryptophan depletion sometimes depend upon (T + treatment) on the first, but not second, day of the variation in associated traits linked to serotonergic func- study (Figure 2). Additionally, the conditional probability tion. Specifically, manipulation of l-tryptophan has been of a cooperative response given a mutually cooperative found to alter reactive aggression in participants with high, outcome on the previous round of the game was similarly but not low, trait aggression (Cleare and Bond, 1995; Bjork and significantly diminished (Figure 3). Consequently, et al, 2000) and in participants vulnerable to alcoholism these findings suggest that temporary reduction of (LeMarquand et al, 1999; Marsh et al, 2002). Such findings central serotonin actitivity, achieved by manipulation are consistent with evidence, cited above, that aggressive of l-tryptophan, impairs the ability of healthy adults to behavior depends upon serotonergic mechanisms in experi- learn (or acquire) a pattern of cooperative behavior in the mental, clinical, and naturalistic settings (Raleigh et al, context of a validated laboratory measure of reciprocal 1991; Virkkunen et al, 1994; Giovanni et al, 2001 for a review). However, the present data also demonstrate We acknowledge that the above pattern of data may an effect of tryptophan depletion on the acquisition of reflect the influence of uncontrolled factors such as family cooperative behavior in a laboratory measure of social history of mood disorder and, in the female participants, cooperation that appears largely independent of variation phase of menstrual cycle that might have interacted with in physical or verbal aspects of trait aggression, its tryptophan depletion to compromise cooperative behavior (common) emotional correlate (anger), or the possession while playing the PD game. However, importantly, the of critical social attitudes (as indexed by trait hostility).
reduced cooperative responses associated with the TÀ treat- One notable feature of our results is that there was a ment on the first study day are not attributable to, or significant interaction between treatment and treatment secondary to, temporary changes in mood. Consistent with order such that tryptophan depletion significantly affected previous studies (Moore et al, 2000), our participants, who performance of the PD game on the first but not the second did not include any with a history of mood disorder, did study day. Similar interactions have been reported fre- not report significantly greater reductions in state posi- quently in pharmacological studies of cognitive function tive affect, or increases in state negative affect, after the that have used within-subject, crossover designs, including Effects of tryptophan depletion on PD game performanceRM Wood et al Table 2 Subjective Effects for Participants Who Drank an Amino-Acid Drink Containing l-Tryptophan (T+) and Participants Who Drankan Amino-Acid Drink not Containing l-Tryptophan (TÀ) on Study Day 1 and Day 2 State positive and negative affect (PANAS). Visual analog scales for state aggression (VAS). * ¼ F(1,20) ¼ 5.83, po0.05.
Effects of tryptophan depletion on PD game performance some involving tryptophan depletion (Park et al, 1994).
cingulate cortex and anteroventral striatum compared to all These effects reflect the fact that drug treatments can have other outcomes of the game (ie mixed cooperation/ different effects depending upon whether tasks are com- defection and mutual defection). That data suggest that pleted for the first or second time and are particularly mutual cooperation activates neural systems of the fore- marked where the tasks involve significant learning (Coull brain, perhaps supporting the development of social et al, 1995; Elliott et al, 1997). In the present study, analysis reciprocity over repeated encounters (Rilling et al, 2002, of the simple effects indicated that tryptophan depletion 2004). Evidence has continued to accumulate suggesting reduced participants’ cooperative behavior on the first that serotonin plays a significant role in modulating the study day by approximately 31%. This impression was reinforcement processes mediated by this circuitry and its reinforced by pair-wise comparisons indicating that parti- mid-brain dopaminergic innervation. This evidence in- cipants who underwent the TÀ treatment on study day 1, cludes demonstrations that serotonergic activity enhances and who showed relatively reduced number of cooperative intracranial self-stimulation (Redgrave and Horrell, 1976) responses, significantly increased their cooperative re- and that SSRI treatment decreases self-stimulation thresh- sponses after the T + treatment on study day 2, while olds (Harrison and Markou, 2001). Additionally, potentiat- participants who underwent the T + treatment on day 1 ing and diminishing serotonergic activity has been found to showed similarly high levels of cooperation on day 2.
increase and decrease the reinforcing effects of cocaine, Evidence indicates that the tit-for-tat strategy followed by respectively (Aronson et al, 1995; Sasaki-Adams and Kelley, the fictional partner of the PD game in the present study 2001). Finally, Rogers et al (2003) recently demonstrated (actually a prepared computer program) is highly effective that tryptophan depletion altered healthy participants’ in eliciting cooperation from other playing partners processing of prospective gains, but not prospective (Sheldon, 1999). Consequently, these data suggest that losses or probability cues, when choosing between binary- temporary disruption of serotonin impairs the acquisition outcome gambles, suggesting that serotonin can, under of cooperative behaviors even when the strategies of social appropriate conditions, modulate risky choice through the partners are geared to elicit reciprocal cooperation.
control of attention towards appetitive signals. Therefore, Serotonin has also been shown to influence learning our finding that tryptophan depletion reduces cooperative mechanisms that may mediate acquisition of socially choices on an iterated PD game of the kind employed in the cooperative behavior (Fletcher et al, 1999; Sasaki-Adams present study suggests that serotonin modulates activity and Kelley, 2001) and which depend upon much the same within circuitry encompassing the rostro-cingulate cortex neural circuitry as performance of an iterated PD game and the striatum that mediates the reinforcing effects of (Rilling et al, 2002). Serotonergic depletions within the cooperative behavior (Rilling et al, 2002). This hypothesis forebrain of monkeys (Clarke et al, 2004), and tryptophan may provide a basis for understanding why major depletion in humans (Park et al, 1994; Rogers et al, 1999), depressive disorder is associated with social withdrawal impair the ability to learn and relearn changing stimulus- (see Fossi et al, 1984; Weissman, 2000) by positing a failure reinforcement associations, probably reflecting altered of serotonergic innervation to modulate appropriately the neuromodulation of circuitry encompassing the orbito- reinforcement value of social interactions (Libet and frontal cortex, cingulate cortex, and striatum (Cardinal Lewinsohn, 1973; Youngren and Lewinsohn, 1980).
et al, 2002). In the former study with healthy human Additionally, it is notable that the reported effects were participants (Park et al, 1994), tryptophan depletion also apparent in an iterated, sequential PD game in which the had a disproportionately greater impact on such associative participants of the study made their choice about whether learning on the first study day when participants completed to cooperate or defect only after their playing partnerF the task for the first time. The capacity to link (and in fact, a computer playing a strict tit-for-tat strategyF dynamically) relink stimuli to motivationally relevant had indicated his/her response. In this case, cooperative outcomes is critical for effective function within social responses involved foregoing higher rewards available from environments in which such linkages are apt to change a defection response (ie 30p) in preference for smaller rapidly and unpredictably (Rolls, 1996). Such conditions rewards available from mutually cooperative responses (ie clearly obtained on study day 1 on which participants were 20p). Behavioral analyses of iterated PD games have shown required to assess the behavior of their partner and to that a player has two incompatible objectives in these adjust their own strategies in the light of varying outcomes circumstances (Raichlin et al, 2001). On the one hand, he/ from each round of the PD game. These conditions were she wishes to defect; on the other hand, he/she wishes to get less apparent on day 2 when participants repeated the task his/her partner to cooperate. However, their partner’s strict with the same fictional (confederate) and real (computer) tit-for-tat strategy ensures that any defection will incur a retaliatory defection on the immediately following round of These results shed light on the underlying role of the game. At this point, the player will face a choice between serotonin in mediating social cooperation in healthy human becoming locked in an ongoing pattern of mutual defec- participants, and highlight several directions for future tions or be forced to cooperate (while his partner defects) in research into this area, in addition to suggesting candidate order to re-establish a pattern of mutual cooperation. In this neuropsychological mechanisms for social dysfunction in way, playing the iterated, sequential PD game, in which the partner plays strict tit-for-tat, involves the exercise of self- Brain-imaging studies have reported that mutual co- control in forgoing the higher immediate reward produced operative outcomes in the context of an iterated PD game of by defecting when the partner has cooperated in favor of a kind similar to that played in this study was associated the immediately smaller (but cumulatively greater) rewards with increased BOLD signal within the rostral anterior associated with enduring mutual cooperation.
Effects of tryptophan depletion on PD game performanceRM Wood et al Consistent with work indicating that tit-for-tat strategies Bechara A, Tranel D, Damasio H, Damasio AR (1996). Failure to cooperative behavior (Axelrod, 1984; Sheldon, 1999), the T respond automatically to anticipated future outcomes following + treatment in the current study produced a clear damage to prefrontal cortex. Cerebral Cortex 6: 215–225.
preference for cooperation on both the first (0.7370.06) Biggio G, Fadda F, Fanni P, Tagliamonte A, Gessa G-L (1974).
and second day (0.6470.09), suggesting that performance Rapid depletion of serum tryptophan, brain tryptophan, was supported by a strong representation of the longer-term serotonin and 5-hyrdoxyindoleacetic acid by a tryptophan-freediet. Life Sci 14: 1321–1329.
benefits of mutual cooperation. Serotonin has frequently Bjork JM, Dougherty DM, Moeller FG, Swann AC (2000).
been associated with aspects of impulsivity in both clinical Differential behavioral effects of plasma tryptophan depletion (Soubrie, 1986; Virkkunen et al, 1994) and experimental and loading in aggressive and nonaggressive men. Neuropsycho- settings (Dalley et al, 2002), with demonstrations that tryptophan depletion can impair impulse control in healthy Bond A, Lader M (1974). The use of analogue scales in rating controls (Walderhaug et al, 2002) and in individuals with a subjective feelings. Br J Med Psychol 47: 211–218.
family history of alcoholism (Crean et al, 2002; LeMarquand Boone C, de Brabander B, Carree M, de Jong G, van Olffen W, van et al, 1999). Other data have suggested that depletion of Witteloostuijn A (2002). Locus of control and learning to cooperate central serotonin impairs the capacity to delay gratification in a prisoner’s dilemma game. Person Indiv Differ 32: 929–946.
(Denk et al, 2005; Mobini et al, 2000; Wogar et al, 1993).
Brown GL, Goodwin FK, Ballenger JC, Goyer PF, Major LF (1979).
Aggression in humans correlates with cerebrospinal fluid amine Consequently, the present findings suggest that tryptophan metabolites. Psychiatry Res 1: 131–139.
depletion significantly weakens the influence of the longer- Buss AP, Perry MP (1992). The aggression questionnaire. J Person term rewards associated with mutual cooperative strategies, perhaps represented within circuitry encompassing the Cardinal R, Parkinson J, Hall J, Everitt B (2002). Emotion and orbitofrontal cortex (Bechara et al, 1996); thereby, under- motivation: the role of the amygdala, ventral striatum, and mining the basis for sustained reciprocation. This hypo- prefrontal cortex. Neurosci Biobehav Rev 26: 321–352.
thesis may provide a basis for understanding how Carpenter LL, Anderson GM, Pelton GH, Gudin JA, Kirwin PDS, impulsiveness in personality disorders, associated with Price LH et al (1998). Tryptophan depletion during continuous serotonergic dysfunction, undermines social function CSF sampling in healthy human subjects. Neuropsychopharma- through an inadequate representation of the future Clarke HF, Dalley JW, Crofts HS, Robbins TW, Roberts AC (2004).
reward-value of reciprocal cooperation. It also predicts Cognitive inflexibility after prefrontal serotonin depletion.
that serotonergic manipulations will be most potent in determining social interactions, such as those involving Cleare AJ, Bond AJ (1995). The effect of tryptophan depletion tit-for-tat interactions, which pit immediate and delayed and enhancement on subjective and behavioural aggression in prospects for reinforcement against each other; by contrast, normal male subjects. Psychopharmacology 118: 72–81.
serotonergic manipulations in the context of other PD Coccaro EF, Kavoussi RJ (1997). Fluoxetine and impulsive games involving other kinds of playing strategies may not aggressive behavior in personality-disordered subjects. Arch In summary, we have demonstrated that depriving Coull JT, Middleton HC, Robbins TW, Sahakian BJ (1995).
healthy participants of l-tryptophan reduced cooperative Contrasting effects of clonidine and diazepam on tests of workingmemory and planning. Psychopharmacology 120: 311–321.
behavior and the tendency of mutual cooperation to Crean J, Richards JB, de Wit H (2002). Effect of tryptophan promote further cooperation. These results provide a depletion on impulsive behaviour in men with or without a starting point for testing novel hypotheses concerning the family history of alcoholism. Behav Brain Res 136: 349–357.
role of serotonin in social cooperation and its dysfunction Dalley JW, Theobald DE, Eagle DM, Passetti F, Robbins TW (2002).
in psychiatric disorders. These include (i) a failure to find Deficits in impulse control associated with tonically-elevated cooperation with others adequately reinforcing and (ii) serotonergic function in rat prefrontal cortex. Neuropsycho- weakening the strength of long-term gains (associated with mutual cooperation) over short-term immediate rewards Denk F, Walton ME, Jennings KA, Sharp T, Rushworth MF, (associated with defection following cooperative responses Bannerman DM (2005). Differential involvement of serotonin and dopamine systems in cost–benefit decisions about delay oreffort. Psychopharmacology 179: 587–596.
Dubini ABM, Polin V (1997). Do noradrenaline and serotonin differentially affect social motivation and behaviour? European Neuropsychopharmacology 7(Suppl 1): S49–S55.
This research was funded by an independent award from Elliott R, Sahakian BJ, Matthews K, Bannerjea A, Rimmer J, the Biotechnology and Biological Sciences Research Council Robbins TW (1997). Effects of methylphenidate on spatialworking memory and planning in healthy young adults.
(BBSRC; UK) to Robert D Rogers (S20137).
First MB, Spitzer RL, Gibbon M, Williams JBW (2002). Structured Clinical Interview for DSM-IV-TR Axis I Disorders, ResearchVersion, Patient Edition. (SCID-I/P). Biometrics Research: New Aronson S, Black J, McDougle C, Be S, Jatlow P, Heninger G et al (1995). Serotonergic mechanisms of cocaine effects in humans.
Fletcher PJ, Korth KM, Chambers JW (1999). Selective destruction of brain serotonin neurons by 5,7-dihydroxytryptamine in- Axelrod R (1984). The Evolution of Cooperation. Penguin Books: creases responding for a conditioned reward. Psychopharma- Axelrod R, Hamilton WD (1981). The evolution of cooperation.
Fossi L, Faravelli C, Paoli M (1984). The ethological approach to the assessment of depressive disorders. J Nerv Ment Dis 172: 332–341.
Effects of tryptophan depletion on PD game performance Fuller RW (1996). The influence of fluoxetine on aggressive influenced behaviors in vervet monkeys (Cercopithecus aethiops behavior. Neuropsychopharmacology 14: 77–81.
sabaeus). Psychopharmacology 72: 241–246.
Giovanni PA, Placidi MA, Oquendo MA, Malone KM, Huang YY, Redgrave P, Horrell RI (1976). Potentiation of central reward by Ellis SP et al (2001). Aggressivity, suicide attempts, and localised perfusion of acetylcholine and 5-hydroxytryptamine.
depression: relationship to cerebrospinal fluid monoamine metabolite levels. Biol Psychiatry 50: 783–791.
Rilling JK, Gutman DA, Zeh TR, Guiseppe P, Berns GS, Kilts CD Harrison A, Markou A (2001). Serotonergic manipulations both (2002). A neural basis for social cooperation. Neuron 35: potentiate and reduce brain stimulation reward in rats: involvement of serotonin-1A receptors. J Pharmacol Exp Ther Rilling JK, Sanfey AG, Aronson JA, Nystrom LE, Cohen JD (2004).
The neural correlates of theory of mind within interpersonal Higley JD, Mehlman PT, Poland RE, Taub DM, Higley SB, Suomi SJ interactions. Neuroimage 22: 1694–1703.
et al (1992). Cerebrospinal fluid monoamine and adrenal Robbins TW, Cador M, Taylor J, Everitt B (1989). Limbic-striatal correlates of aggression in free-ranging rhesus monkeys. Arch interactions in reward-related processes. Neurosci Biobehav Rev Higley JD, Mehlman PT, Poland RE, Taub DM, Vickers J, Suomi SJ Robbins TW, Everitt BJ (1996). Neurobehavioural mechanisms of et al (1996). CSF testosterone and 5-HIAA correlate with different reward and motivation. Curr Opin Neurobiol 6: 228–236.
types of aggressive behaviors. Biol Psychiatry 40: 1067–1082.
Rogers RD, Blackshaw AJ, Middleton HC, Matthews K, Deakin Knutson B, Panksepp J (1996). Effects of fluoxetine on play JFW, Sahakian BJ et al (1999). Tryptophan depletion impairs dominance in juvenile rats. Aggress Behav 22: 297–307.
stimulus-reward learning while methylphenidate disrupts atten- Knutson B, Wolkowitz OW, Cole SW, Chan T, Moore EA, Johnson RC tional control in healthy young adults: implications for the et al (1998). Selective alteration of personality and social behaviour monoaminergic basis of impulsive behaviour. Psychopharma- by serotonergic intervention. Am J Psychiatry 155: 373–379.
LeMarquand DG, Benkelfat C, Pihl RO, Palmour RM, Young SN Rogers RD, Tunbridge EM, Bhagwagar Z, Drevets W, Sahakian BJ, (1999). Behavioral disinhibition induced by tryptophan deple- Carter CS (2003). Tryptophan depletion alters the decision- tion in nonalcoholic young men with multigenerational family making of healthy participants through altered processing of histories of paternal alcoholism. Am J Psychiatry 156: 1771–1779.
reward cues. Neuropsychopharmacology 28: 153–162.
Libet J, Lewinsohn PM (1973). The concept of social skill with Rolls ET (1996). The orbitofrontal cortex. Philos Trans Roy Soc special references to the behaviour of depressed persons.
J Consult Clin Psychol 40: 304–312.
Sasaki-Adams DM, Kelley AE (2001). Serotonin–dopamine inter- Marsh DM, Dougherty DM, Moeller FG, Swann AC, Spiga R (2002).
actions in the control of conditioned reinforcement and motor Laboratory-measured aggressive behavior of women: acute behavior. Neuropsychopharmacology 25: 440–452.
tryptophan depletion and augmentation. Neuropsychopharma- Sheldon KM (1999). Learning the lessons of tit-for-tat: even competitors can get the message. J Person Soc Psychol 77: Mobini S, Chiang TJ, Ho MY, Bradshaw C, Szabadi E (2000).
Effects of central 5-hydroxytryptamine depletion on sensitivity Soubrie P (1986). Reconciling the role of serotonin neurons in to delayed and probabilistic reinforcement. Psychopharmacology human and animal behaviour. Behav Brain Sci 9: 319–362.
Trivers R (1971). The Evolution of Reciprocal Altruism. Q Rev Biol Moore P, Hans-Peter L, Seifritz E, Clark C, Bhatti T, Kelsoe J et al (2000). Clinical and physiological consequences of rapid Trivers R (1985). Social Evolution. Benjamin/Cummings: Menlo tryptophan depletion. Neuropsychopharmacology 23: 601–622.
Moskowitz DS, Cote S (1995). Do interpersonal traits predict Tse WS, Bond AJ (2002a). Serotonergic intervention affects both affect? A comparison of three models. J Person Soc Psychol 69: social dominance and affiliative behaviour. Psychopharmacology Moskowitz DS, Pinard G, Zuroff DC, Annable L, Young SN (2001).
Tse WS, Bond AJ (2002b). Differences in serotonergic and The effect of tryptophan on social interaction in everyday life: a noradrenergic regulation of human social behaviours. Psycho- placebo-controlled study. Neuropsychopharmacology 25: 277–289.
Nelson HE (1982). National Adult Reading Test (NART) Test Virkkunen M, Rawlings R, Tokola R, Poland RE, Guidotti A, Nemeroff C et al (1994). CSF biochemistries, glucose metabo- Park SB, Coul JT, McShane RH, Young AH, Sahakian BJ, Robbins lism, and diurnal activity rhythms in alcoholic, violent offenders, TW et al (1994). Tryptophan depletion in normal participants fire setters, and healthy participants. Arch Gen Psychiatry 51: produces selective impairments in learning and memory.
Walderhaug E, Lunde H, Nordvik JE, Landro NI, Refsum H, Pruitt D, Kimmel M (1977). Twenty years of experimental gaming: Magnusson A (2002). Lowering of serotonin by rapid trypto- critique, synthesis, and suggestions for the future. Annu Rev phan depletion increases impulsiveness in normal individuals.
Raichlin H, Brown J, Baker F (2001). Reinforcement and punish- Watson D, Clark LA, Tellegen A (1988). Development and ment in the Prisoner’s Dilemma game. Psychol Learn Motiv 40: validation of brief measures of positive and negative affect: The PANAS scales. J Person Soc Psychol 54: 1063–1070.
Raleigh MJ, Brammer GL, McGuire MT, Yuwiler A (1985).
Weissman MM (2000). Social functioning and the treatment of Dominant social status facilitates the behavioral effects of depression. J Clin Psychiatry 61(Suppl 1): 33–38.
serotonergic agonists. Brain Res 348: 274–282.
Wogar MA, Bradshaw CM, Szabadi E (1993). Effect of lesion of the Raleigh MJ, McGuire MT, Brammer GL, Pollack DB, Yuwiler A ascending 5-hydroxytryptaminergic pathways on choice between (1991). Serotonergic mechanisms promote dominance acquisi- delayed reinforcers. Psychopharmacology 111: 239–243.
tion in adult male monkeys. Brain Res 559: 181–190.
Youngren MA, Lewinsohn PM (1980). The functional relation Raleigh MJ, Yuwiler A, Brammer GL, McGuire MT, Geller E, between depression and problematic interpersonal behavior.
Flannery JW (1981). Peripheral correlates of serotonergically-

Source: http://www.rogerslab.org.uk/downloadWood_et_al_2006.pdf