Music discriminations by carp (Cyprinus carpio)

Fish live in a medium that is an especially good conductor of sound, and the myriad extant species—some 25,000 in the class Osteichthyes (bony fishes) alone (Nelson, 1984)—have evolved diverse mechanisms for transducing acoustic stimuli. It should not be surprising if sophisticated functionality in the auditory domain turns out to be a major component of the perceptual and cognitive capabilities of fish. This paper’s point of departure is the elegant study by Porter and Neuringer (1984), which demonstrated the hitherto unsuspected capability of pigeons to classify music into stylistic categories, just as humans do. My interest in testing koi on a similar task evolved into the present series of experiments, which range over the areas of discrimination learning, categorization, reversal learning, and music perception and which indicate the existence of hitherto unsuspected capabilities in fish. Like Porter and Neuringer (1984), I prefer to optimize the chance of discovering currently unknown capabilities in an animal by seeking a positive result in a top-down fashion, first challenging the subject with a task complex enough to require such capabilities, and then simplifying only as necessary. The current understanding of fish psychoacoustics does nothing to discourage high expectations of fishes’ higher perceptual functions. In the case of perhaps the most extensively studied species, “absolute detection thresholds, frequency discrimination, intensity discrimination, temporal summation, complex spectrum discrimination, temporal discrimination and resolution, and various measures of ‘auditory filter’ characteristics show that the goldfish falls solidly within a general vertebrate pattern of auditory processing” (Fay & Ream, 1986, p. 1883). Furthermore, goldfish are known to perceive acoustic dimensions equivalent to what humans perceive as pitch and timbre (Fay, 1995). Indeed, for more than a decade, it has seemed clear that humans and animals share the basic psychoacoustic functions and that there are probably no fundamental qualitative differences in auditory perception between humans and at least those fish whose anatomy allows them to hear reasonably well (Fay, 1992, 1994, 1995, 1998; Jacobs & Tavolga, 1968; Popper & Fay, 1993; Popper, Platt, & Saidel, 1982). The present subjects were koi, which are members of the carp family and close relatives of the goldfish. Neither goldfish nor koi are known to communicate by making sounds (Fay, 1995), but both are classified as otophysans, a group that has bony structures (the Weberian ossicles) coupling the swim bladder to the inner ears, in which there are hair cells with a specialized orientation pattern. Because of this anatomy, carp are among the species referred to as hearing specialists, which have the greatest auditory bandwidth and sensitivity of any fish (Popper & Fay, 1993). Figure 1 shows the frequency sensitivity range of koi in comparison with the spectral ranges of other animals and of music. Their relatively low upper limit of hearing means that koi would hear music roughly as it would sound to us over an ordinary voice-grade telephone line. In one of the more complex experiments in fish psychoacoustics, Fay (1992) conditioned goldfish to a two-tone chord and then tested them on a range of individual pitches, obtaining essentially a bimodal generalization curve whose peaks corresponded to the component pitches of the training chord. Thereby, Fay was the first to demonstrate analytic listening (Hartman, 1988) in a nonhuman animal—in this case, the ability to discriminate the individual components This work was supported by the Rowland Institute for Science. I gratefully acknowledge the help of Winfield Hill for instrumentation, programming, and encouragement. I thank Dan Coutu for the MIDI music and Matt Maltzman for ongoing discussions and help with the manuscript. Thank you, Angel Peterchev, for suggesting Paganini’s theme for the melody experiment. Thanks are due Chris Stokes and the Rowland Staff. Special thanks to Allen Neuringer for his inspiration and invaluable comments. Correspondence concerning this article should be addressed to A. R. Chase, Rowland Institute for Science, 100 Edwin Land Boulevard, Cambridge, MA 02142 (e-mail: [email protected]) . Music discriminations by carp (Cyprinus carpio)