Learning from the behaviors and experiences of others

Abstract

Learning to fear and avoid what is dangerous is crucial for survival. Perhaps equally important is the ability to learn that something which was previously dangerous is now safe. Although we can learn about dangers individually, through our own experiences, it is likely more safe to learn about them from others, by observing their behaviors and reactions. In a sense, this allows us to learn through the experiences of others. The overarching goal of this thesis is to deepen our understanding of how we learn about fear and safety through observation of others.

In Study I we let participants undergo an observational extinction paradigm to investigate if safety learning was facilitated through observation of a calm learning model. In a direct conditioning stage participants first learned to associate a stimulus with fear. Next, they learned through that the previously feared stimulus was now safe. This extinction of fear was either direct or vicarious (observational). We demonstrated that attenuation of fear was greater following vicarious rather than direct extinction. We further showed that this was driven by the learning model’s experience of safety.

Although learning through others is likely an efficient way of learning, observational learning also has to be applied critically, for instance by not copying the choices of someone that performs poorly. In Study II and Study III we investigated how people learned to make choices through observation of others, demonstrators, which had either a high or low ability. In both studies, participants learned a simple probabilistic two-choice task to avoid shock. Results from Study II demonstrated that people were able to use the observational information to improve performance regardless of the ability (skill) of the demonstrator. They only copied the choices of the demonstrator with high ability and they were able to learn from observing the consequences of a demonstrator’s choice regardless of the demonstrator’s ability. In Study III we also provided participants with descriptions of the abilities of the demonstrators. Our results showed that describing the demonstrator as low in ability impaired observational learning, regardless of the actual ability of the demonstrator and that this is likely driven by a difference in attention directed towards the observational information.

An inability to discriminate threatening from safe stimuli is typical for individuals suffering from anxiety. In Study IV we investigated how observational fear conditioning is affected by the learning model’s expressed anticipatory anxiety. Results showed that participants were able to discriminate the threatening from the safe stimuli equally well from a learning model that behaved anxiously (i.e. did not discriminate) as from one that did not behave anxiously (i.e. did discriminate).

The results presented in this thesis increase our understanding of how healthy individuals learn about aversive events and stimuli through observation of the behaviors and reactions of others and how these reflect the observed individuals’ experiences.