Do chimpanzees seek explanations? Preliminary comparative investigations

Canadian Journal of Experimental Psychology, Jun 2001 by Daniel J Povinelli, Sarah Dunphy-Lelii

Videotape coding. All trials from the testing phase of this and all subsequent studies were videotaped from two perspectives for coding. A set of standardized, written instructions was administered to a main rater, who coded all trials (N = 112), and a reliability rater, who coded a random sample of two of each of the subject's trials (n = 56). The instructions asked the raters to record the following information: (1) "Did the chimp inspect a block with his/her fingers and/or bring the block towards themselves to smell, taste, and/or closely visually inspect the block?" (2) "Did the chimp switch a block to the opposite [platform] and attempt to stand it up?" (3) "Did the chimp succeed in standing one or both blocks at any point during the trials?" For any positive answers, the raters indicated which block or blocks were relevant. Cohen's kappas (kappa) of 1.00, 0.84, and 1.0, were obtained for each question, respectively. The data from the main rater was used for all analyses. In addition, a main rater and a reliability rater separately recorded the duration of time the apes spent manipulating the sham block on each probe trial. Pearson's correlation yielded a coefficient of determination, r^sup 2^, of 0.998.

RESULTS

Throughout testing, the apes were highly interested and motivated to stand the blocks correctly. First, and as expected, the apes succeeded in standing up 99% of the functional blocks during the testing sessions (in contrast to only a single case in which one of the apes (Apollo) successfully stood up a sham block by balancing it against the edge of the mat which covered the surface of the platform). Second, the apes exerted considerable effort attempting to stand the sham block. Figure 2a displays the mean percent of time that the apes spent trying to stand the sham block in blocks of two trials. Averaged across trials 1-2, the subjects spent 44 (SD = 21.6) of the 120 seconds attempting to stand the sham blocks, versus 23.2 (SD = 16.8) averaged across trials 3-4, a decline that was statistically significant, t(6) = 6.551, p

The main results of the study are presented in Table 1, which shows the percent of functional and sham blocks examined by each subject. There was not a single case in which subjects examined the functional blocks. In contrast, on average, the subjects examined 17.9% of the sham blocks. (Because the subjects never examined the functional blocks, a direct statistical comparison between the frequency of functional and sham block examinations was not possible.) More importantly, perhaps, five of the seven subjects (71%) were each coded as having examined one sham block. Two of these examinations occurred in session 1, two occurred in session 2, and one occurred in session 4.

The five instances of block inspections were microanalyzed to provide a more detailed description of the inspection process. It was difficult to distinguish between close visual inspection and sniffing. Thus, two categories were delineated: sniffing/close visual inspection and visual inspection alone. Four of the apes engaged in an instance of sniffing/close visual inspection of the blocks (Apollo, Mindy, Candy, Kara), and two apes exhibited an instance of close visual inspection alone (Kara, Megan). Megan also exhibited a bout of backing up and visually scanning between both blocks as they sat on their respective platforms. Kara exhibited a single bout of visually examining the bottom of the functional block after she had previously successfully stood it upright. Interestingly, this occurred immediately after an unsuccessful attempt to stand the sham block. Mindy was the only ape who (at least arguably) engaged in a bout of tactile inspection; she put the block into her mouth after her bout of sniffing/close visual inspection.