Pepperberg (parrot learning) Flashcards
what is the title of this study?
acquisition of the same/different concept by an African grey parrot: learning with respect to categories of colour, shape and material
what year was this study conducted?
1987
what was the psychology being investigated in this study?
comprehension of same and different
what was the background of this study?
to see if nonhumans can use abstract symbolic relationships when communicating
many psychologists believe that only humans possess “true language skills” alongside the ability to show a range of cognitive skills
prior to this study, pepper berg had reported on an african grey parrot, Alex
he could categorise subjects, count uo to six and use functional phrases +come here” “I want ____” “wanna go ____” and “no”
however, pepper berg stated that these do not show whether a non-human can comrehend and use abstract symbolic relationships when communicating
one cognitive skill that had been reported as being a concept not seen in non-humans is the comprehension of “same” or “different”
premark noted that for a non-human to demonstrate comprehension of “same” two aspects must apply
they must recognise that two independent objects called A1 and A2 are both blue and this single attribute makes them “same”
they must also recognise that this sameness can be immediately extrapolated and symbolically represented not only for two other blue items, but for two novel independent items that have nothing in common with the original set of A’s
this study was designed to test these two way for assessing the cognitive skill of “same” or “different”
what were the aims of this study?
to see if an avian subject could use vocal labels to demonstrate symbolic comprehension of the concepts of same and different
what was the procedure of this study?
research method: lab experiment, case study
experimental design: N/A
IV: whether the object is familiar or novel
DV: whether the parrot responds correctly to the question
Sample: African grey parrot called Alex who was the focus of pepper bergs work since June 1977. he had free access to all parts of the lab for 8 hours a day when trainers were present. during his sleeping hours he was placed in a cage with water and a standard seed mix for parrots was available at all times. the trials occurred at various locations around the laboratory depending on where Alex was at the time. other food such as fresh fruits, vegetables, nuts and toys were provided when Alex asked for it
sampling technique: opportunity sampling
qualitative data: % success rate on trials was measured for familiar and novel objects
Alex was presented with two objects which could be differentiated used on three categories: colour, shape, and material. he would then be asked what’s same or what’s different
a correct response would only be recorded if Alex vocalised the appropriate category
four processes Alex had to go through to get a correct response
attend to multiple features of two different objects
from the vocal question, determine whether the response was based on sameness or difference
work out what is same or different
vocally produce a category response
to complete these, Alex had to perform the cognitive skill of feature analysis on the objects
he had already been learning language and concepts for around 9 years prior to the study hence he could already produce vocal labels in English
during the course of the study, Alex acquired labels for orange, purple, and sex cornered objects
training sessions occurred 2-4 times a week and lasted between 5 to 60 minutes
how was the general training conducted in this study?
MODEL/RIVAL technique - the primary technique used by pepper berg. it is based on the principles of social learning theory
this demonstrates to the parrot types of interactive responses in the study
one person acts as a trainer to the second human. the trainer asks questions about the object and gives praise and rewards for the correct answer but shows disapproval for the incorrect answer
the second human acts as a model for Alex but also as a rival for the trainers attention
the roles of model/rival and trainer were frequently reversed and Alex was often given the opportunity to participate in these sessions
during any training where the purpose was to acquire a correct label , each correct response was rewarded with the object itself
to keep Alex’s motivation high, he could ask for any reward if he answered correctly
how was same/different training conducted in this study?
a trainer would hold up two objects in front of the model/rival and ask “whats same” or “what’s different”
both types of questions and training objects were mixed with each session
objects were always red, green or blue, triangular or square
model/rival would respond with the correct category label and was given a reward
if the m/r gave an incorrect response, the trainer scolded the person
when an error was presented, the objects were removed from sight, and then presented again with the same question asked
the role of m/r was then reversed
initial training contrasted just the categories of colour and shape Alex had already learnt. he was then trained on a third category “man-mah/ matter”
to prevent boredom of repetitive tasks he was also being trained on number concepts, new label for other objects, recognition of photographs and object permanence
formal testing was only started after he acquired the label “Mah-mah”
how was the 2nd general training conducted?
a secondary trainer who had never trained him earlier carried out trials. this was done to reduce any effect of cuing from the original trainers
the questioning was incorporated into other test sessions that were being conducted on Alex
on a previous day, the principal trainer would list all possible objects that could be used for testing. a student who was not involved in any training would then choose the question and randomly order them
in a week, same or different questions were asked 1- 4 times
testing took place over 26 months
principal trainer was presented wherever the trail took place but she sat with her back facing Alex and did not look at him during the presentation of objects
she never knew what was being presented and would repeat out what she thought Alex had said
if it was correct Alex was given that object and would be praised
if not, the examiner removed the object and emphatically said “NO!” when this happened, a correction procedure was used in which the objects presented until the correct response was given
the same materials were never presented again so there was a sings first trial response
an overall test score was produced. first trial results were also calculated
what were the tests on familiar objects in this study?
objects paired were presented to Alex
they were similar pairings to the ones used in the training phase but never the same
individual objects were obviously used in more than one trial but the pairing were always novel and a specific pair would only ever be presented again if Alex gave an incorrect label
what was the transfer tests using novel objects in this study?
Alex was presented with pairs of objects that might not have even had a label
he was exposed to objects that did not have a label for and objects that he has no experience of
any completely new object was within the environment of Alex for several days prior to being used so that Alex got used to seeing it and to reduce fear responses
how were probes used in this study?
one concern was that in formulating his answers, Alex might not be attending to the questions, but merely responding to the physical characteristics of the objects
thus, at random intervals probes were administered in which he was asked questions for which two category labels could be the correct response
if he were ignoring the content of the question and answering on the basis of attributes, he would have responded with an incorrect answer
what were the results of these studies?
the training for Alex to acquire colour and shape as labels took 4 months and for “man-mah” it took 9 months.
the length of each session was dictated by Alexis willingness to attend
what were the results for familiar objects?
99 out of 126 correct responses overall 76.7%
69 out of 99 on fist trial only performance 69.7%
based on chance, he should have scored 33%
his performance on pairs made of objects that were no longer novel, bu contained a colour, shape or material he could not yet label was 13 out of 17 and 10 out of 13 for first trials
what were the results for transfer tests with novel objects?
96 out of 113 correct overall 85%
79 out of 96 on fist trials 83.2%
when there was a novel object in a pair his sore was 86% and when both objects were novel it was 83%
