Oddity discrimination of category in the honeybee

Date

2022

Contributor

Department

Instructor

Depositor

Speaker

Researcher

Consultant

Interviewer

Narrator

Transcriber

Annotator

Journal Title

Journal ISSN

Volume Title

Publisher

University of Hawaii at Manoa

Volume

Number/Issue

Starting Page

Ending Page

Alternative Title

Abstract

Studies of associative learning in bees demonstrate remarkably similar findings to those obtained with vertebrates. More recent work has begun to examine more complex relational learning phenomena. Giurfa (2001) found same/different learning in honeybees using a Y- maze procedure, and Muszynski and Couvillon (2015) obtained the first evidence of oddity learning in honeybees. Further research by Muszynski and Couvillon (2020) found that category difference facilitated trial-unique oddity learning in honeybees. Vertebrates struggle with oddity learning problems, with even the best-performing species (apes and monkeys) taking hundreds or thousands of trials to solve the problems, and other species (cats, rats, raccoons) failing to solve the problems altogether. Therefore, it is very impressive that an invertebrate like the honeybee demonstrates oddity learning. Following Muszynski and Couvillon (2020), the present experiment is a further exploration of category difference in oddity problems. Honeybees (Apis mellifera) were trained using the free-flying method, in which individuals were recruited from sugar water feeders and brought to an experimental chamber. The chamber contained a stimulus with high-concentration sucrose; the bee drank that sucrose and then returned to the hive to deposit it. Due to the high concentration of sucrose, the bee was then motivated to return to the chamber for more. Each trial consisted of one visit to the chamber. The design of Experiment 1 was a three-stimulus oddity problem in which choice of the odd stimulus was rewarded. Two categories of stimuli were used: circles of a single color (solid) and circles composed of two differently colored semicircles (split). The colors used were green (G), orange (O), yellow (Y), and blue (B). Thus, there were 4 unique solid stimuli and 6 unique split stimuli. The training was trial-unique, such that each trial had a new configuration of stimuli. On half the trials a solid stimulus was odd, and on the rest a split stimulus was odd. The two trial types were intermixed over the training trials. For Group Identical, the nonodd stimuli were exactly the same (e.g., O-/O-/YB+, BG-/Y+/BG-). For Group Nonidentical, the nonodd stimuli were different from each other (e.g., O-/B-/YB+, BG- /Y+/BO-). The expectation was that Group Identical bees would find the problem easier; they could choose the odd stimulus on the basis of color, category, or both. Group Nonidentical bees could choose the odd stimulus only on the basis of category. Both groups solved the problem, with no difference in either rate of learning or asymptote. The results indicate that honeybees can learn to choose the odd stimulus on the basis of category, which is a novel finding; in previous work category had been confounded with other dimensions such as color. Experiment 2 was designed to examine bees’ ability to discriminate solid and split stimuli outside of an oddity context. The design featured four stimuli separated into two pairs by a wooden partition. One pair consisted of solid stimuli while the other consisted of split stimuli, and each bee was consistently rewarded for choosing the pair from one of the two categories. For bees in Group Identical, the two stimuli within a pair were exactly the same (e.g., YO YO | G G). For bees in Group Nonidentical, the two stimuli within a pair were different (e.g., BY GY | B O). As in Experiment 1, the training was trial-unique such that bees could not solve the problem by learning about specific stimuli. Bees in both groups solved the category discrimination, and there was no difference between the groups in rate of learning or asymptote. Therefore, it appears that honeybees can discriminate solid and split stimuli and use this category discrimination as a tool for solving oddity problems. Future work might examine the use of category in a different type of relational learning problem, e.g. a same/different discrimination. Overall, this research shows evidence of simple category discrimination as well as category discrimination in an oddity problem. Discrimination is unaffected by additional variation in the stimuli (i.e. whether pairs of stimuli are identical or not). These findings are remarkable given that vertebrates have relative difficulty with oddity problems. Further studies of the relational learning of honeybees are needed to determine the generality of these conceptual capabilities.

Description

Keywords

Honeybee, Discrimination learning

Citation

Extent

Format

Geographic Location

Time Period

Related To

Related To (URI)

Table of Contents

Rights

Rights Holder

Local Contexts

Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.