162
diffi cult for the researcher to control the playback stimuli. This is a major problem
when trying to determine auditory thresholds to acoustic playback stimuli or in
behavioral tasks that involve sound-level discrimination. Second, operant conditioning methods require signifi cant motivation of the animal being tested; feeding
assays are only effective in cases where the reinforced reward has a strong biological signifi cance to the animal (e.g., a food reward to a food-restricted animal). This
may signifi cantly impact the training time required for learning. For example, using
this method, goldfi sh can be trained within 1–2 days, whereas oscars ( A. ocellatus )
can take 12–14 days to train (Yan and Popper 1992 ). Finally, the operant conditioning paradigm requires the use of an unconditioned stimulus that is biological relevant to the animal’s unconditioned response. In the case of a food reward, the
unconditioned stimulus would require modifi cation of a feeding refl ex. In the Yan
and Popper ( 1991 ) conditioning paradigm, this required a species specialized for
strike-feeding with the striking motion being conditioned to the auditory stimulus.
Since not all fi sh have the same mode of prey capture, this may make a similar operant conditioning paradigm diffi cult to use in comparative studies.
2.1.1 Avoidance Conditioning
Avoidance conditioning is a variant of the operant conditioning paradigm that has
been successfully used to determine the audiograms of a number of fi sh species.
Tavolga and Wodinsky ( 1963 ) were the fi rst to perform an exhaustive comparative
study of the auditory capacities of nine species of fi shes using avoidance
Fig. 2 Cross-section diagram of the apparatus used for operant conditioning by Yan and Popper
( 1992 ) to determine the hearing capability of the oscar ( Astronotus ocellatus ). Fish were trained to
press the observation paddle (O) and attend to a sound played through an underwater speaker (S).
If they perceived a sound stimulus, they were trained to press the response paddle (R). Correct
responses were rewarded with a food pellet delivered from an automatic feeder (F) through a tube
(T). Incorrect responses resulted in the ceiling light (C) being switched off. Modifi ed from Yan and
Popper ( 1992 )
A.A. Bhandiwad and J.A. Sisneros
Précédent

- 174/488

Suivant