Orbitofrontal cortex contributes to the comparison of values underlying economic choices

Economic decision-making hinges on the ability to assess and compare the subjective values of different options. While previous studies have identified neuronal correlates of value representation and choice in the orbitofrontal cortex (OFC) of monkeys, the precise role of the OFC in the comparison of values remains unclear. Several neural substrates have been proposed, including motor systems, distributed processes, shifts of visual attention, hippocampal integration, and even models that don't involve explicit value comparison. However, these hypotheses lack causal support. This study aimed to investigate whether the OFC causally contributes to the comparison of values underlying economic choices. The researchers developed a novel paradigm to assess the causal role of a neural population in a decision process. A binary decision is essentially a comparison of two neural signals, such as offer values in economic choice. The experiment involved manipulating neural activity and measuring choice variability. Increased variability could stem from noisier input signals, a noisier decision process, or motor response disruption. By controlling for the first and third possibilities, the researchers could infer the involvement of the neural population in the decision process itself. Previous research highlighted some relevant findings. Low-current stimulation of the middle temporal area biased decisions, while high-current stimulation increased choice variability, primarily due to noisier input signals. Optogenetic inactivation of OFC in mice increased choice variability due to stereotyped behavior, indicating OFC's necessity for economic choices but not specifying whether its role is in value assignment, comparison, or both. Previous experiments by the authors showed that electrical stimulation of OFC during offer presentation altered offer values and, at high currents, increased choice variability. This was potentially due to value comparison disruption but could have also involved other brain regions or motor planning disruption. The current study aimed to specifically test whether weak electrical stimulation of OFC increases choice variability without inducing valuation biases, providing causal evidence for OFC's role in value comparison.

The understanding of neural mechanisms underlying economic decision-making has evolved significantly. Early work by Padoa-Schioppa and Assad (2006) demonstrated that neurons in the OFC encode economic value. Subsequent research by Padoa-Schioppa (2013) explored the neuronal origins of choice variability in economic decisions. Other studies have investigated the role of various brain regions and processes in economic decisions, including the motor systems (Song et al., 2017; Zhang et al., 2018), distributed processes (Glimcher et al., 2005; Cisek, 2007; Cisek, 2012; Hunt & Hayden, 2017), shifts in visual attention (Krajbich et al., 2010), hippocampal signals (Bakkour et al., 2019), and models that do not explicitly compare values (Hayden & Moreno-Bote, 2018). However, many of these studies are correlational, making it difficult to establish causal relationships.

Two male rhesus monkeys underwent surgery for head-restraining devices and recording chambers. Tungsten electrodes were implanted bilaterally in the central orbital gyrus, targeting area 13/11 of the OFC. Monkeys performed a two-alternative forced choice task where they selected between two juices offered sequentially in variable amounts, aiming to discourage premature decisions. Weak electrical stimulation (5-15 µA), differing from previous high-current stimulation experiments, was delivered during either the first or second offer. Stimulation parameters were tightly controlled (timing, duration, pulse characteristics, frequency). The monkeys' choices were analyzed using probit regression to assess the relative value (indifference point), sigmoid steepness (inversely related to choice variability), and order bias (preference for the first or second offer). Several control analyses were performed to account for potential confounding factors, such as changes in motivation, stereotyped behavior (reversion to side bias, choice hysteresis based on juice type, side, or order), and error rate, all assessed through additional probit regression models.

Weak electrical stimulation (5-15 µA) of OFC during the presentation of the second offer, but not the first, significantly increased choice variability. This finding indicates that OFC contributes specifically to value comparison, as opposed to value representation or motor planning. The stimulation did not significantly alter the relative value, order bias, or introduce stereotyped behavioral patterns (side bias, choice hysteresis). Furthermore, the stimulation did not increase the error rate or alter response times, ruling out changes in motivation or task engagement as explanations for the observed increased variability. The researchers compared the effect sizes for different levels of electrical stimulation (5-15 µA, 25 µA, 50 µA, and 125 µA), pooling data from this study and a previous one. They observed a U-shaped trend for the increase in choice variability: significant effects at very low and high currents, but not intermediate ones. This suggests different cellular mechanisms underlying the variability increase at low versus high currents. The authors hypothesized that weak stimulation might disproportionately increase inhibition, disrupting the excitation-inhibition balance necessary for accurate value comparison. The absence of effects at intermediate currents might reflect balanced excitation and inhibition, while high-current stimulation might involve additional mechanisms, such as antidromic spikes or fiber of passage disruption. Importantly, the effect size for choice variability was greater at 5-15µA than at 125 µA.

The results provide causal evidence that neurons in the primate OFC participate in value comparison during economic choices. The use of weak electrical stimulation, combined with the absence of spatial signals in the OFC, strongly suggests that the stimulation did not affect action planning directly. The lack of effect on offer values (relative value and order bias) supports the specificity of the effect on value comparison. The finding that weak stimulation selectively affects value comparison during the second offer but not during the first further reinforces this conclusion, as the second offer is where the comparison takes place. The results are consistent with good-based models of decision-making, which posit that value comparison is partially or entirely in good-based representations, and contradict action-based models that imply decision-making occurs in motor areas through competition between action plans. Similarly, the results contradict sequential decision models (attentional drift diffusion model and accept/reject models), which would predict effects from stimulation during either offer.

This study provides definitive causal evidence for the role of the OFC in value comparison during economic decision-making in primates. The findings demonstrate that OFC neurons are causally involved in both the computation and comparison of offer values. Future research should focus on the circuitry within the OFC and its connections with other brain regions to further elucidate the mechanisms of value comparison and the complex interaction of excitation and inhibition.

The study was conducted on only two monkeys, limiting the generalizability of the results. While the researchers carefully controlled for various potential confounding factors, the possibility of other unmeasured factors influencing the results remains. Furthermore, the study focused specifically on value comparison; the findings do not necessarily exclude the contribution of other brain regions to this process.