Thursday, June 30, 2016

"Pessimistic rats are extra sensitive to negative feedback"

I don't know if it's traders or politicians that got me thinking along these lines but the BPS is on to something by pointing this out.

From the British Psychological Society's Research Digest blog:

Depression is complex and influenced by many factors, but the way depressed people think is a likely contributor to the disorder. Depression is often associated with cognitive biases, including paying more attention to negative than positive events and recalling them more easily. People with depression also tend to ruminate over perceived failures and criticism, and they are extra sensitive to negative feedback.

Analogous cognitive biases can be found in animals. Now, in a new study, researchers have demonstrated for the first time a link between pessimism and sensitivity to performance feedback in rats. It’s the latest finding to show parallels between human depression and rat pessimism – an important result that lends further legitimacy to using animal research to shed light on human psychological problems.

You might wonder how on earth it is possible to measure pessimism in rats.  One way to do this was shown in a 2004 study in which animals were trained to press a lever to receive a food reward in response to hearing one tone, and to press a different lever to avoid a mild electric shock upon hearing a different tone. Then the scientists presented the animals with intermediate tones, in between the ones that signaled either food or shock. Which lever the rats pressed in response to these ambiguous cues was considered an indicator of whether the animals expected a positive or negative event. In other words, their behaviour revealed their relative optimism or pessimism.

In the new research, Rafal Rygula and Piotr Popik of the Polish Academy of Sciences used the same paradigm to compare the reaction of rats displaying optimistic and pessimistic traits to positive and negative feedback. First, they divided rats into two groups based on how they performed in the ambiguous-cue interpretation test. Some rats tended to interpret ambiguous cues as signaling a reward, indicating a positive cognitive bias, while others were more likely to interpret them as signaling punishment, indicating a bias toward more pessimistic judgments.

Then the optimistic and pessimistic rats were trained and tested in a probabilistic reversal-learning (PRL) task, which essentially involves using negative or positive feedback to teach the animals to change or maintain a response that they’ve learned previously. Rygula and Popik determined how likely each rat was to switch its response after receiving negative feedback and to maintain its response following positive feedback....MORE