Tryptophan Depletion Promotes Habitual over Goal-Directed Control of Appetitive Responding in Humans

Abstract

Background: Optimal behavioral performance results from a balance between goal-directed and habitual systems of behavioral control, which are modulated by ascending monoaminergic projections. While the role of the dopaminergic system in behavioral control has been recently addressed, the extent to which changes in global serotonin neurotransmission could influence these 2 systems is still poorly understood.

Methods: We employed the dietary acute tryptophan depletion procedure to reduce serotonin neurotransmission in 18 healthy volunteers and 18 matched controls. We used a 3-stage instrumental learning paradigm that includes an initial instrumental learning stage, a subsequent outcome-devaluation test, and a slip-of-action stage, which directly tests the balance between hypothetical goal-directed and habitual systems. We also employed a separate response inhibition control test to assess the behavioral specificity of the results.

Results: Acute tryptophan depletion produced a shift of behavioral performance towards habitual responding as indexed by performance on the slip-of-action test. Moreover, greater habitual responding in the acute tryptophan depletion group was predicted by a steeper decline in plasma tryptophan levels. In contrast, acute tryptophan depletion left intact the ability to use discriminative stimuli to guide instrumental choice as indexed by the instrumental learning stage and did not impair inhibitory response control.

Conclusions: The major implication of this study is that serotonin modulates the balance between goal-directed and stimulus-response habitual systems of behavioral control. Our findings thus imply that diminished serotonin neurotransmission shifts behavioral control towards habitual responding.

Keywords: 5-HT; goal-directed behavior; inhibitory control; serotonin; stimulus-response habit learning; tryptophan depletion.

© The Author 2015. Published by Oxford University Press on behalf of CINP.

Figures

Figure 1.

Instrumental learning task. (A) Instrumental learning stage. In this example, in 2 different trials, participants were presented with cherries and pear on the outside of the box. For each of the trials, if the correct key was pressed (R – right for cherries and L – left for pear), participants were rewarded with another fruit (apple and grapes, respectively) on the inside of the box and points. If the incorrect key was pressed, an empty box was shown and no points were earned. (B) Outcome-devaluation test. In this example, 2 open boxes with an apple and grapes inside were presented to the participants. The fruit (the apple in this example), which was no longer worth any points, was indicated with a red cross on it. The participant should press the correct key for the still-valuable fruit (left key in this example for the still-valuable grapes symbol). (C) Slips-of-action test. In this example, the initial instruction screen showed that the kiwi and the apple outcomes (indicated by the red crosses) will now lead to the subtraction of points. The other 4 outcomes were still valuable. Following the instruction, participants were shown in rapid (2 seconds) succession the fruit stimuli (on the front doors of the boxes) and were asked to press the correct keys (Go-trial) for the still-valuable outcome stimuli (inside the box) but to refrain from responding (No-Go trials) when the outcome stimuli were devalued. In this particular example, participants should withhold responding when the cherry stimulus was shown but press the left key for the pear stimulus. (D) Baseline control test of inhibitory response. In this example, the initial instruction screen showed that 2 of the cueing stimuli on the outside of the boxes indicated by the red crosses (banana and pear) would now lead to the subtraction of points. The other 4 stimuli were still valid. Following the instruction, participants were shown the fruit stimuli in rapid (2 seconds) succession and asked to press the correct keys (Go-trial) for the still-valid stimuli but to refrain from responding (No-Go trials) when the stimuli were devalued. In this particular example, participants should withhold response when the pear stimulus was shown but press the right key for the cherry stimulus.

Figure 2.

(A) Behavioral results in a slip-of-action test: the goal-directed system (stimulus-response-outcome association) allows a successful performance based on action outcome evaluation; the habitual system, based on stimulus-response (S-R) association, activates response directly and will enable slips of action toward no-longer-valuable outcomes. (B) Percentage of correct responses in the instrumental learning stage of the task. (B1-8) The testing blocks of 12 trials each (96 trials in total). (C) Reaction time to the key press in the instrumental learning stage.

Figure 3.

(A) Percentage of response on valuable and devalued stimuli in the slips-of-action test. (B) Percentage of response to correct (Go) and incorrect (NoGo) stimuli in the internal control test of inhibitory response control. (C-D) Regression analysis of percentage of responses to the devalued outcomes out of the total number of trials with devalued outcomes in slip-of-action test over percentage of tryptophan reduction in peripheral blood in acute tryptophan depletion (ATD) and control group (BAL) (C) and BAL (D).