The Brain and Logical Thinking

How does one solve a math problem on an exam that was not in the homework? Say the problem required us to solve a simple definite integral. To solve it we would make an inference: if the antiderivative of f’(x) is this and if the fundamental theorem of calculus states this then the answer must be this. We also readily must make inferences to determine the value of a particular thing. For example, say one must make an inference to determine the worth of attending class: if I’m sick and if the probability of transmission is high then attending class is not worth it. Naturally, the ability to combine different data points and derive conclusions (i.e. inferences) from them is crucial in many other aspects of our lives beyond school.

 

So, what part of the brain enables us to make inferences? For some time now it has been known that the orbitofrontal cortex (OFC) is crucial for our ability to infer. Studies using model organisms have found that animals with an inactivated or damaged OFC were unable to infer the value that a thing has. But animals with an intact OFC were able to. Curiously, however, other studies have shown that animals with an inactivated or damaged OFC could determine a thing’s value by really lying on their experience (as opposed to inference). So, these results indicate that OFC enables us to infer from different pieces of information but how the OFC does so is unclear. Many of these studies used preconditioning tasks to establish that the OFC enables inferential thinking. But whether the OFC only responded to information that was associated with a value was not determined. This led Brain F. Sadacca et al. to investigate if the OFC only uses information that has been associated with a value to make inferences or if it can use any information (regardless of value) to make inferences.

 

Sadacca and his team used single-cell recording on rats to record the activity of various neurons in the OFC during a sensory preconditioning task. In the first stage of sensory preconditioning valueless (i.e. neutral) cue pairs are associated. Then one cue from each pair is associated with a reward. So, if the strong neural activity would occur during the first stage this would indicate that the OFC uses information regardless of that piece of information’s value. Sadacca found that neurons were highly active while the one cue from each pair was associated with a reward, but during the association of the neutral pairs. Thus, as Sadacca wrote, “these results support a role for OFC in representing associative structure, independent of value” (Sadacca, 2018).

 

In a review article written by Fang Wang and Thorsten Kahnt, they discuss different studies that support Sadacca’s findings and that have further expanded what is knowledge about how our brains enable us to make inferences. Like Sadacca, Wang and Kanhnt used a sensory preconditioning task one mice. They found that mice were able to predict rewards after the cue that was not associated with a value was presented, indicating that they were able to inference (as the cue was only associated with a different cue that was associated with a reward). It was also found that the neutral cue pair association came not only from the OFC but also from the hippocampus (HC). The review article mentioned an experiment that used a Pavlovian devaluation task to investigate the OFC's role in our ability to infer the value of a particular thing. It found that the circuitry between the OFC and amygdala played a crucial role in enabling this ability.

 

References

Sadacca, B. F., Wied, H. M., Lopatina, N., Saini, G. K., Nemirovsky, D., & Schoenbaum, G. (2018, March 7). Orbitofrontal neurons signal sensory associations underlying model-based inference in a sensory preconditioning task. eLife. Retrieved October 23, 2021, from https://elifesciences.org/articles/30373.

Wang, F., & Kahnt, T. (2021, March 6). Neural circuits for inference-based decision-making. Current Opinion in Behavioral Sciences. Retrieved October 23, 2021, from https://www.sciencedirect.com/science/article/pii/S2352154621000267?via%3Dihub.