Sensitivity of the brain to loss aversion during risky gambles

doi:10.1016/j.tics.2007.05.006

Trends in Cognitive Sciences

Volume 11, Issue 7, July 2007, Pages 270-272

https://doi.org/10.1016/j.tics.2007.05.006 Get rights and content

Little is known about the neural systems that subserve human loss aversion. A recent neuroimaging study by Tom, Poldrack and colleagues reports that this pattern of behaviour is directly tied to the greater sensitivity of the brain to potential losses compared with potential gains and uncovers a brain network whose activity increases with potential gains and decreases with potential losses. These results challenge the common view that loss aversion engages a distinct emotion-related brain network (e.g. amygdala and insula).

Introduction

When deciding between risky options, humans are about twice as sensitive to the possibility of losing goods or money than to the possibility of winning them. Prospect theory, the leading behavioural model of decision making under risk, uses the concept of loss aversion (as measured by the willingness to reject gambles) to explain risk aversion during monetary mixed (gain or loss) gambles (Box 1) [1]. Several recent brain imaging studies have suggested that higher sensitivity to loss entails emotional processes recruiting structures such as the amygdala and the anterior insula 2, 3, 4, 5. However, the recent study by Tom et al. reported that neither of these two brain regions showed increasing activity with the size of potential losses [6]. In fact, the same neural substrates (e.g. striatum and ventromedial prefrontal cortex) exhibited both decreased activity with potential losses and increased activity with potential gains (Figure 1). Moreover, there was a diminished neural sensitivity to losses among individuals who were less loss averse (i.e. more risk seeking), which might shed light on several neuropsychiatric and behavioural disorders, such as impulsive and risky behaviour, pathological gambling as well as substance abuse. This study elegantly illustrates how the integration of theoretical models and brain imaging approaches provides a better understanding of decision making in risky situations.

Section snippets

The study: deciding between risky options

Tom et al.[6] scanned subjects while they had to accept or reject gambles offering a 50/50 chance of winning an amount of money (range = $10–$40 in $2 increments) or losing another amount ($5–$20 in $1 increments) (Figure 1). All possible combinations of gains and losses were presented. Subjects had to bring $60 on the scanning day and were told that they could actually lose this money in the scanner. Participants pressed one of four buttons to indicate their willingness to play each gamble

Single or different neural substrates for potential gains and loss?

On the one hand, brain regions showing increased activity with potential gains included a ‘gain-brain network’ (striatum, ventromedial prefrontal cortex, anterior cingulate cortex and midbrain) previously observed during anticipation and receipt of monetary gain 7, 8, 9 or rewarding juice [10]. Thus, correlation with the size of potential gains when evaluating gambles without expectation of immediate reinforcement elicits neural responses similar to those observed during anticipation and

Concluding remarks

This study provides important new insights into the functional properties of decision making in humans. The reduced neural sensitivity to losses among individuals who were less loss averse is particularly relevant for several neuropsychiatric and behavioural disorders, such as substance abuse and pathological gambling, associated with increased risk taking and impulsive behaviour. These individual differences in behavioural and neural loss aversion might be related to naturally occurring

Acknowledgements

J-C.D. is partially supported by a European reintegration grant from the 6th Framework Program, the Fyssen and the Medical Research foundations. I thank G. Coricelli, P. Domenech and G. Sescousse for comments on an early version of the manuscript.

References (15)

C.M. Kuhnen et al.
The neural basis of financial risk taking
Neuron
(2005)
H.C. Breiter
Functional imaging of neural responses to expectancy and experience of monetary gains and losses
Neuron
(2001)
B. Knutson
A region of mesial prefrontal cortex tracks monetarily rewarding outcomes: characterization with rapid event-related fMRI
Neuroimage
(2003)
J.P. O’Doherty
Neural responses during anticipation of a primary taste reward
Neuron
(2002)
M.P. Paulus et al.
An insular view of anxiety
Biol. Psychiatry
(2006)
M.V. Peelen et al.
Using multi-voxel pattern analysis of fMRI data to interpret overlapping functional activations
Trends Cogn. Sci.
(2007)
C. Trepel
Prospect theory on the brain? Toward a cognitive neuroscience of decision under risk
Brain Res. Cogn. Brain Res.
(2005)

There are more references available in the full text version of this article.

Cited by (29)

Common and distinct neural substrates of the money illusion in win and loss domains
2019, NeuroImage
Citation Excerpt :
Investigations on the neural circuits involved in the processing of rewards and punishment have produced inconsistent results. On the one hand, a large number of studies have suggested that monetary gains and losses activate a similar fronto-subcortical network, and differ only in degree (Dreher, 2007; Gottfried et al., 2003; Nieuwenhuis et al., 2005; Tom et al., 2007). On the other hand, some studies have indicated that reward and punishment outcomes may involve different neural substrates (Frank et al., 2004; Yacubian et al., 2006).
People often evaluate money based on its face value and overlook its real purchasing power, a phenomenon known as the money illusion. In the present study, using functional magnetic resonance imaging (fMRI) combined with a gambling task, we examined the neural signatures of the money illusion in both win and loss domains. Behavioral results showed that self-reported satisfaction with outcomes was modulated by the face value but not the true value of money in both win and loss domains. At the neural level, activity in the posterior insula was associated with the true value of money in the win domain, but not in the loss domain. Importantly, we found that the ventral striatum, ventromedial prefrontal cortex (vmPFC) and amygdala encoded the money illusion in both domains, indicating a domain-general rather than domain-specific neural signature. Moreover, participants with a larger degree of money illusion at the behavioral level showed stronger functional connectivity between the ventral striatum and ventral anterior cingulate cortex (vACC) in the win domain, but stronger functional connectivity between the ventral striatum and amygdala in the loss domain. Our findings highlight the overlapping and distinct neural substrates underlying the money illusion in the context of wins and losses.
Reduced willingness to invest effort in schizophrenia with high negative symptoms regardless of reward stimulus presentation and reward value
2018, Comprehensive Psychiatry
Citation Excerpt :
Secondly, willingness to invest effort measures captured by our task could be understood as neurobiological markers and, thus, they may be conceptually located closer to the neurobiological basis of the illness rather than the illness phenotype and its resulting behavior. Contrary to the majority of the literature on reward [45], both patients and controls exerted higher effort in gain trials than in loss avoidance trials. Two different theories may explain these apparent counterintuitive results.
Negative symptoms in schizophrenia, which are related to poor functioning, are thought to be grounded on aberrant functioning in the reward system. We aimed to disentangle how negative symptoms and two cognitive aspects of goal-directed behavior, mental representation of reward and reward value, affect willingness to invest effort to attain a reward in schizophrenia.
To this purpose, 43 schizophrenia patients and 35 healthy controls were assessed for negative symptoms and general functioning, and completed an effort-based reward task. Patients were split in high and low negative symptoms scorers. A series of ANOVA tests were conducted in order to test the effects of group controlling for representation of reward (Task 1) and balance between reward value and effort (Task 2) on will to invest effort to attain a reward.
Schizophrenia patients with high negative symptoms chose to invest lower amounts of effort for a reward compared both to low negative symptoms patients and to controls in both tasks. Neither mental representation of reward (Task 1) nor reward value (Task 2) did differentially affect will to invest effort between-groups.
These findings suggest that the lower willingness to invest effort observed in schizophrenia patients with high negative symptoms may not be related to cognitive aspects of goal-oriented behavior.
Beta oscillations and reward processing: Coupling oscillatory activity and hemodynamic responses
2015, NeuroImage
Citation Excerpt :
The Gain > Loss contrast yielded the expected activations in reward-related areas, especially in bilateral ventral striatum and ventromedial PFC and also in bilateral hippocampi (activation in these reward related areas survived a p < 0.001 FWE-corrected threshold at the cluster level; see Fig. 2B and Table 1). At the selected threshold, the Loss > Gain contrast yielded no significant activations, suggesting that a similar brain network is involved in processing both gains and losses but with a differential amount of activation (increasing with gains and decreasing with losses; Dreher, 2007; Tom et al., 2007; Ripollés et al., 2014; Càmara et al., 2009 and Càmara et al., 2010). Thus, this contrast was not further analyzed.
Diverse cortical and subcortical regions are synergically engaged during reward processing. Previous studies using time–frequency decomposition of Electroencephalography (EEG) data have revealed an increase of mid-frontal beta oscillatory activity (BOA) after reward delivery, which could be a potential mechanism in the coordination of the different areas engaged during reward processing. In order to evaluate this hypothesis, twenty subjects performed a monetary gambling paradigm in two separate sessions (EEG and fMRI). Time–frequency oscillatory EEG data and fMRI activity were fused using Joint Independent Component Analysis (ICA). The present results showed that mid-frontal BOA elicited by monetary gains is associated with the engagement of a fronto–striatal–hippocampal network previously involved in reward-related memory enhancement, supporting the role of this activity during reward processing.
Altered cingulo-striatal function underlies reward drive deficits in schizophrenia
2015, Schizophrenia Research
Citation Excerpt :
Our behavioral findings of non-advantageous positive over negative reinforcer correspond to prior findings of selective deficits in positive reinforcement learning in schizophrenia and incentive motivational deficits in first-episode psychosis (Waltz et al., 2007; Murray et al., 2008a). We also observed a decline in Reward Approach accuracy during Pre-reversal due to anticipated loss in healthy controls, which conforms to the theory of loss aversion (Dreher, 2007). The absence of Reward Uncertainty-related accuracy decline in patients is consistent with a prior study demonstrating a failure in adjusting value according to the prospect of loss in schizophrenia (Trémeau et al., 2008).
Amotivation in schizophrenia is assumed to involve dysfunctional dopaminergic signaling of reward prediction or anticipation. It is unclear, however, whether the translation of neural representation of reward value to behavioral drive is affected in schizophrenia. In order to examine how abnormal neural processing of response valuation and initiation affects incentive motivation in schizophrenia, we conducted functional MRI using a deterministic reinforcement learning task with variable intervals of contingency reversals in 20 clinically stable patients with schizophrenia and 20 healthy controls. Behaviorally, the advantage of positive over negative reinforcer in reinforcement-related responsiveness was not observed in patients. Patients showed altered response valuation and initiation-related striatal activity and deficient rostro-ventral anterior cingulate cortex activation during reward approach initiation. Among these neural abnormalities, rostro-ventral anterior cingulate cortex activation was correlated with positive reinforcement-related responsiveness in controls and social anhedonia and social amotivation subdomain scores in patients. Our findings indicate that the central role of the anterior cingulate cortex is in translating action value into driving force of action, and underscore the role of the cingulo-striatal network in amotivation in schizophrenia.
Evaluation of an operant successive negative contrast task as a method to study affective state in rodents
2012, Behavioural Brain Research
Citation Excerpt :
An individual's sensitivity to reward gain or loss has been proposed to be influenced by emotional state [1]. In general people are more sensitive to possible reward losses than gains [2] and evidence shows that the behavioural and neurophysiological responses of people in a negative affective state show an enhanced sensitivity to loss or failure [3]. Increased sensitivity to loss appears to be symptomatic of negative affective states in people and it is hypothesised that biases in information processing and underlying mechanisms relating to the evaluation of reward gains and losses, may reliably reflect affective state in animals [4].
Successive negative contrast (SNC) describes a change in an animal's behaviour following a downshift in the quantitative or qualitative value of a reward. Previous studies suggest both consummatory and instrumental paradigms have the potential to provide an objective measure of affective state in rodents. We first investigated whether an SNC effect is observed in an operant task based on the 5 choice serial reaction time task. We then tested whether this SNC effect was sensitive to differences in affective state induced by manipulating the home cage environment. In animals trained to receive a four pellet food reward, reinforcer downshift to a single reward pellet induced a significant slowing of both correct response and collection latencies to levels below that of animals which had only ever received the lower value reward, indicating a SNC effect. Home cage environmental enrichment resulted in a paradoxical effect on responses in this SNC task where animals housed in a barren environment showed faster baseline response times and the SNC effect was significantly attenuated. These data suggest that the animals housed in the barren conditions were in a more positive affective and/or motivational state during testing than animals housed in enriched cages. Although opposite to the effects of housing conditions in a runway SNC task, these data could be explained by the enriching effects of daily training in an operant task. Rather than inducing a negative affective state in rats, the barren housing conditions resulted in a relatively more positive affective state in the chamber when compared to animals living in a highly enriched environment.
Functional Neuroimaging of Reward Circuitry Responsivity to Monetary Gains and Losses in Posttraumatic Stress Disorder
2009, Biological Psychiatry
Citation Excerpt :
Numbing of responsiveness to positive stimuli and anhedonia are common in PTSD patients (1) and are considered by some to be the most specific diagnostic features of the disorder (36). The pattern of results obtained in the healthy subjects (viz., striatal activation to gains accompanied by striatal deactivation to losses) is consistent with prior observations (21,37–43). Interestingly, the present PTSD subjects showed not only less striatal activation to gains but also a trend toward less striatal deactivation to losses (which would have been statistically significant had this result been predicted a priori).
Clinical impressions and preclinical work suggest that posttraumatic stress disorder (PTSD) might be associated with dysfunctional reward processing. To pursue this issue, we administered a validated passive-viewing monetary reward task during functional magnetic resonance imaging (fMRI) to subjects with chronic PTSD and to mentally healthy individuals.
The protocol evaluated fMRI signal changes that anticipated or accompanied monetary gains and losses under varying conditions of controlled expectation. The “expectancy phase” entailed presentation of a promising, unpromising, or intermediate Wheel of Fortune–type spinner, whereas the “outcome phase” was defined by the arrow landing on one of three sectors of that spinner, thereby determining the subjects' gain or loss for that trial.
Neuroimaging data from 20 PTSD and 26 healthy subjects withstood quality control procedures and were included. In voxelwise and anatomically defined region-of-interest analyses, when gains were contrasted to losses, between-group comparison revealed smaller bilateral striatal activations in the PTSD subjects. In the PTSD group, less striatal activation to gains versus losses was associated with more self-reported motivational and social deficits.
The present data support the hypothesis that PTSD is associated with abnormal processing of monetary outcomes and that this alteration might be related to some aspects of emotional numbing.

View all citing articles on Scopus

View full text

Trends in Cognitive Sciences

UpdateResearch FocusSensitivity of the brain to loss aversion during risky gambles

Introduction

Section snippets

The study: deciding between risky options

Single or different neural substrates for potential gains and loss?

Concluding remarks

Acknowledgements

Neuron

Neuron

Neuroimage

Neuron

Biol. Psychiatry

Trends Cogn. Sci.

Brain Res. Cogn. Brain Res.

Update
Research Focus
Sensitivity of the brain to loss aversion during risky gambles