Please enable JavaScript.

Coggle requires JavaScript to display documents.

Social Learning - Coggle Diagram

- - - - test phase - trials where hand on screen same as one pp's performing (compatible) vs trials where pp's opening hand diff to what on screen (incompatible)
      - training phase - ctrl group had compatible stimulus-response training, experimental = incompatible stimulus response training
      - same no. of responses/stimuli but DIFFERENCE in contingency of stimulus (sensory perception) and motor responses
      - FOUND can learn NOT to imitate
        
        incompatible training abolished automatic imitation effect (this was where ppl had slower RT's for incompatible trials than compatible ) - shows assosciative learning principles CAN change imitation
        
        BUT most adults can imitate most actions and difficult actions are difficult to measure RT - more practical to look at learning NOT to imitate so we can conclude learning theory principles play role
    - - does imitation depend on contingency
        
        ie Hammond 1980 - varied prob. of water delivery when no lever pressing - same contiguity (temporal relationship) but low contingency (water also delivered when lever not pressed) -- led to LOWER responding
        
        contingency grp - received CONTINGENT incompatible stimulus-response training - even if incompatible, presence of stimulus predicted response made
        
        non-contingent grp of non-contingent incompatible stimulus-response training - had to make a response even in absence of stimuli
        
        contingent grp showed MORE learning and less imitation than the non-contingent grp - quicker RT - less automatic imitation
      - is imitation context-dependant
        
        peak and bouton's rsch - when rats reintroduced to context a (tone > shock) even after learned ctxt b (tone > food) - they freezed anticipating the shock - ctxtl cues are stimuli themselves that insert in assosciation
        
        vs rescorla-wagner model that would suggest surprise of ctxt b would mean unlearn tone assosciated with shock
        
        Cook tested imitation in contexts A and B - blue and red rooms - gave them incompatible stimulus-response training in only one of the ctxs and tested imitation again in BOTH ctxts
        
        more learning (less imitation) seein in CONTEXT where had incompatible training - imitation IS ctxt dependant
- - - - fMRI showed humans learn to attend to social input using SAME assosciation processes relating to paying attention to non-social
  - - - learning increases when unpredictable/uncertain outcomes
      - mirrors high prediction error = lots of learning
    - - learning increases in proportion to no. of times relationship btwn events is experienced
      - greater contingency?
    - - learning increases when rewarded
- - - - enjoyment - similarity NEGATIVELY predicted and contingency POSITVELY predicted
      - closeness to others ( picking one of venn diagrams) - contingency POSITIVELY predicted these scores
      - Helping - pp's in contingent grp more likely to show up to follow up
- - - - active of process learning through observation - person creating symbolic rep of bhvr guiding future bhvrs
      - 3 factors in modelling
        
        characteristics of model - like similarity, simple, aggression/hostility
        
        attributes of observer - ie low confidence/self esteem (copy when uncertain rule) - history of conforming (copy the majority)
        
        consequences of imitating bhvr (copy the successful) - ie positive results, link to conditioning
    - - opportunities to UNLEARN things to - ie video games to promote prosocial behaviour
        
        ppl who played prosoc video games (harrington and O Connell) scored higher on empathy measures and on prosoc bhvr even when ctrlling for video games