Gossette & Hood: Discrimination Reversal, Variation of Motivational and Incentive Levels

Question 1

Background

Answer 1

Successive discrimination reversal (SDR) is one of the most promising approaches to evaluating performance differences among species

Differences found in 9 species of birds and 6 species of mammals

With birds: SDR data led to a rank of species that were correlated w/both assumed phyletic differences and indices of brain complexity

Differences in SDR among closely related birds (ie: intrafamily) were less than remotely related birds

Motivation and incentive were undefined from the experiments

Experiments used only sufficient food deprivation and food reward to make sure a response would occur

Difference in SDR between 2 birds was so disparate that it led researchers to believe that motivational level may be a critical variable

A certain amount of excitement or “emotionality” impaired the reversal performances

“emotionality” may be viewed as motivational variable

Question 2

Method

Answer 2

This study compared SDR scores for a high and low drive level (D) and measured the effect of incentive (K) at each drive level

4 pigeons in 4 groups with D and K values in a formboard box

2 levels of D had 10% and 30% reduction of feeding weights

2 levels of K with 1 and 4 seeds/response

Training to make sure that there was no preference by the pigeons for any one group

Pre-training and then Preference testing and forced responding sessions to counterbalance response to both block positions

Question 3

Results and Discussion

Answer 3

Incentive had a significant effect upon total errors, total sessions, but NOT total initial errors

Drive level had a significant effect on total initial errors but NOT total errors or total sessions

While drive and incentive are sources of control of the rate of error reduction in SDR trials, only incentive has a persistent and important role

Maybe b/c K works on a superordinate habit such as “learning set or reversal set” that is responsible for a decrease in errors across problems

Major source of intergroup differences were to due to differences in incentive

Only incentive significantly contributed to asymptotic differences with high incentive leading to uniformly lowest error scores

Low drive level led to highest latency

Significant elevation of initial error by high drive implies that locus of drive is in the increased resistance to extinction of the previously correct response; supported by past research

Question 4

Learning set (LS) methodology

Answer 4

Ignores the drive variable except to provide sufficient food deprivation

One of the most popular measures of LS is the 2nd trial response (shortened measure of initial error)