Environmental enrichment of brown capuchins (Cebus apella): Behavioral and plasma and fecal cortisol measures of effectiveness

Abstract

No consensus exists about the quantity and variety of environmental enrichment needed to achieve an acceptable level of psychological well-being among singly housed primates. Behavioral and plasma and fecal cortisol measures were used to evaluate the effectiveness of four levels of toy and foraging enrichment provided to eight wild-caught, singly housed adult male brown capuchins (Cebus apella). The 16-week-long study comprised six conditions and began with a 4-week-long preexperimental and ended with a 4-week-long postexperimental period during which the subjects were maintained at baseline enrichment levels. During the intervening 8 weeks, the subjects were randomly assigned to a sequence of four 2-week-long experimental conditions: control (baseline conditions), toy (the addition of two plastic toys to each cage), box (access to a foraging box with food treats hidden within crushed alfalfa), and box & toy (the addition of two plastic toys and access to a foraging box). Behavioral responses to changes in enrichment were rapid and extensive. Within-subject repeated-measure ANOVAs with planned post hoc contrasts identified highly significant reductions in abnormal and undesirable behaviors (and increases in normal behaviors) as the level of enrichment increased from control to toy to box to box & toy. No significant behavioral differences were found between the control and pre- and postexperimental conditions. Plasma and fecal cortisol measures revealed a different response to changing enrichment levels. Repeated-measure ANOVA models found significant changes in both these measures across the six conditions. The planned post hoc analyses, however, while finding dramatic increases in cortisol titers in both the pre- and postexperimental conditions relative to the control condition, did not distinguish cortisol responses among the four enrichment levels. Linear regressions among weekly group means in behavioral and cortisol measures (n = 16) found that plasma cortisol was significantly predicted by the proportions of both normal and abnormal behaviors; as the proportion of normal behaviors increased, the plasma cortisol measures decreased. Plasma cortisol weekly group means were also significantly and positively predicted by fecal cortisol weekly group means, but no behavioral measure significantly predicted fecal cortisol weekly group means. In sum, these findings argue strongly that access to a variety of toy and foraging enrichment positively affects behavioral and physiological responses to stress and enhances psychological well-being in singly housed brown capuchins. Am. J. Primatol. 48:49–68, 1999. © 1999 Wiley-Liss, Inc.