W9: Associative Learning Flashcards
(23 cards)
What is associative learning?
A connection or association formed between two stimuli or between a behaviour and a response
Describe the procedure of classical conditioning
US -> a stimulus that is naturally occurring and triggers a response e.g. food
UR -> natural response to an US
CS -> neutral initially but after being paired with an US it can elicit a response
CR -> learned response to a conditioned stimulus e.g. salivation in response to a bell
What is latent inhibition?
Describes the delayed learning of a conditioned response to a stimulus that has previously been encountered without any significant consequence- it illustrates how prior exposure to a stimulus can affect an individuals’ ability to learn new associations with a stimulus
Stimulus does not predict any outcome so prior exposure makes it harder to later associate the stimulus with a new response
Emphasises the importance of past experiences on current learning processes
How has latent inhibition been observed in rats?
When rats are pre exposed to a tone without any other consequence, they are slower to learn that the tone will predict a reward than those not exposed
Easier to find in animals but harder in humans
What is perceptual learning?
Exposure to or experience with a set of stimuli generated from the same prototype defined category that can enhance our performance when we are called to recognise those stimuli- mechanism that leads to this improvement is known as perceptual learning e.g. allows an expert bird watcher to distinguish different birds from each other
How is perceptual learning linked to the inversion effect?
If we pre expose individuals to a category of novel artificial non mono-oriented stimuli e.g. checkerboards, this will enhance their ability to distinguish between two exemplars drawn from that cateogry
What can the basis of the inversion effect for stimuli drawn from a familiar prototype defined category be explained by?
The MKM model of perceptual learning- specifically, the model predicts that its elements that are relatively unpredicted by other elements present that will be salient
How did McLaren, Leevers and Mackintosh (MKM) (1997) investigate perceptual learning of checkerboards?
McLaren showed that we can obtain a signif inversion effect for sets of checkerboards drawn from a familiar category analogous to that in faces- subjects were exposed to a set of checkerboards and were asked to categorise them into two different categories, followed by a discrimination task which included two pairs of checkerboards from a familiar category- results showed that familiarity with a category gave subjects an enhanced ability to discriminate between exemplars of that category in an upright orientation
How did Civile and Zhao (2014) extend the work of McLaren to the face inversion effect?
Extended findings to an old/new recognition task-
show that the inversion effect can be obtained in a standard old/new recognition memory paradigm
Advantage for upright exemplars drawn from a familiar, prototype-defined category, and show that there is a disadvantage for inverted exemplars drawn from this category
Also provided evidence that there is an N170 event-related potential signature for this effect.
Results allow us to integrate a theory of perceptual learning with explanations of the face inversion effect, first reported by Yin.
How was latent inhibition observed in children by Kaniel and Lubow (1986) and what did they find?
Simple exposure/study phase task: children had to press buttons in response to pictures of plants and animals on metal cords in a box divided into 3: cards were presented in sets of three, with one animal card and one plant card on each side of the third card
During each trial, the cards on either side of the middle card were changed and the child had to press a button corresponding to the side on which, e.g. the plant was present
Following the test phase- children were presented with sets of cards corresponding to a card depicting a square they had previously seen in the study phase
They found that children aged 4-5 exhibited reliably poorer learning compared to controls that had not been pre exposed to the test sitmuli, and this effect disappeared in older children (7-10 year olds)
How did McLaren et al (2021) build on Kaniel and Lublow’s (1986) work on latent inhibition in children?
104 primary school children and 32 undergraduate students were used
A pre exposure in study phase of 120 trials where ppts had to respond to clipart pictures preceded by a warning signal (to guard against the possibility of latent inhibition leading to perceptual learning). Only two of the stimuli were used to provide a warning that the next animal was about to be displayed (pre exposed stimuli)- each warning stimulus appeared equally and they then had to complete a task detailing if they had been pre-exposed
What did McLaren et al (2021) find in their work on latent inhibition in children?
Strong evidence for delayed learning as a consequence of preexposure in the youngest group of children, and this decreases with age
Effect on ‘go’ trials has been present in older children, but it is most seen in younger children
This shows that young children display latent inhibition as a consequence of simple pre exposure to stimuli
What are the key features of Skinner’s boxes?
Levers or buttons
Reinforcement mechanism e.g. pressing a lever delivering pos or neg reinforcement
Controlled environment- box allows for the controlling of specific environments
Recording dedvices
How did Morris (1981) measure associative learning and spatial navigation in rats?
Used a Morris pool to test spatial navigation and how a rat explores its environment to form associations with positively or negatively linked stimuli
How did Rodriguez et al (2010) measure rat’s sex differences in spatial navigation using the Morris maze in experiment ONE?
Exp 1- landmark X was a plastic beach ball
Pretraining- placing a rat into the circular pool without landmark but with hidden platform present, which they had to find and stay on for 30 seconds
Training- similar to pretraining but landmark X was always present as well as two boards with triangular shape
Test days- landmark and triangular shaoe were presented 180 degrees apart- amount of time rat spent in front of the areas was recorded
How did Rodriguez et al (2010) measure rat’s sex differences in spatial navigation using the Morris maze in experiment TWO?
Exp 2- landmark X was a ninepin (less salient)
Same process as exp 1 with the different landmark, except the test day was followed by a retraining day and two additional test days
What did Rodriguez et al (2010) find when measuring rat’s sex differences in spatial navigation using the Morris maze in their two part experiment?
When rats were tested without the platform in the presence of the landmark and the shape, a clear sex difference appeared
Males tended to use the geometric info given by the shape of the pool whereas female rats relied more on the landmark
Therefore, experiments 1 and 2 showed that males and females differ in their preference for using a landmark or geometry when solving a spatial task
Further tests in exp 2 revealed that both males and females had learned to find the platform using both sources of info- when the landmark and shape were tested alone, the results showed that both males and females had learned something about their less preferred source of info
How did Civile and Chamizo et al (2014) measure configural information in experiment ONE of rat’s navigation task?
Investigated the effect that disruption of landmark configurations has on rats’ learning in a water maze
Focused on how configural disruption, implemented by flipping the location of both near and far landmarks, significantly affects rat’s performance
Circular swimming pool used with false landmarks
Training consisted of 13 days where the rat was placed in the pool and given 2 mins to find the platform and once found could stay on for 30 secs
Manipulation was to flip the location of the landmarks
What did Civile and Chamizo et al (2014) find when measuring configural information in experiment ONE of rat’s navigation task?
Disrupting configural info caused the rat’s performance to decline (test 1)
A major role that near landmarks play in performance- significant reduction when these were flipped but not the case when ‘far’ landmarks were flipped (test 2)
Swapping the location of the near landmark with the far one disrupts performance compared to the control (test 3)
Across the three tests, sex differences found were due to this sample of female rats not performing as well as the males on the control condition
How did Civile and Chamizo et al (2014) measure configural disruption in their second experiment into rats’ navigation?
Training phase same as exp 1 except there was a directional cue ‘z’- the aim of this was to give the rats a clear distal landmark behind where the platform would be placed- all four landmarks were rotated leaving the directional cue ‘Z’ in the same place as it was in training
What did Civile and Chamizo et al (2014) find in their second experiment into rats’ navigation?
Rats are affected by inversion of the landmark configuration when searching for the platform (test 1)
Analysis on the Z quadrant showed that the time spent in the quadrant was, if anything, lower than expected on a chance basis
‘Far’ landmarks are not particularly important for navigation under these circumstances (test II)
No sex differences were found
What did Civile and Chamizo et al (2020) find in EXPERIMENT 1A when they extended and replicate their initial findings to support the use of configural information in rat’s navigation tasks?
Experiment 1a- showed a strong performance in rats tested with the landmark configuration they were trained on- performance in the inverted condition was signif higher than it was in inverted condition strongly suggests rats were using all of B, C and Z to navigate platform quadrant. f they were using just Z, performance in the flipped condition would be similar to that in the normal condition, and if using B and C, inverted condition would produce similar performance to normal- hence, can conclude that b,C and Z are either being used separately or in configuration
What did Civile and Chamizo et al (2020) find in EXPERIMENT 1B when they extended and replicate their initial findings to support the use of configural information in rat’s navigation tasks?
Location and flipped conditions both show a preference for the Z quadrant over the opposite quadrant. Identity condition can be seen as simply the four landmarks, A, B, C, and D, in the correct orientation with respect to one another but without Z being present.
Implication is that the control exerted by the configuration of landmarks is not that strong in these circumstances and is not having a great impact on behavior. If B and C were acting relatively independently of Z, or as a configuration not involving Z, then we would expect better performance here.
Evidence that Z is capable of being an effective cue on its own (performance in the location condition) and that it does not seem to play a part in configurations with other landmarks (comparison of location and flipped conditions). But we also know from our analysis of Experiment 1a that B and C are also influencing performance because of the significant difference between flipped and inverted conditions in terms of time spent in the Z and BC quadrants
