Abstract
In birds, there is evidence that adult cognitive traits can both run in families and be affected by early developmental influences. However, different studies use different cognitive tasks, which may not be measuring the same traits, and also focus on different developmental factors. We report results from a study in which we administered multiple cognitive tasks (autoshaping, discrimination learning, reversal learning, progressive ratio schedule, extinction learning and impulsivity) to a cohort of 34 European starlings, Sturnus vulgaris, for which several early developmental measures were available. The cohort consisted of siblings raised either apart or together, whose position in the size hierarchy of the rearing brood had been experimentally manipulated. We examined how the different cognitive measures covaried, the extent to which they ran in families, and which of the developmental factors predicted which of the cognitive outcomes. We found that discrimination and reversal learning speeds were positively correlated, as were breakpoint on the progressive ratio schedule and resistance to extinction. Otherwise, the cognitive measures were uncorrelated, suggesting that they reflected different underlying traits. All traits except discrimination and reversal learning speed ran in families to a substantial extent. Using a model selection approach, we found evidence that natal brood size and developmental telomere attrition (the extent to which the birds' erythrocyte telomeres shortened in early life, an integrative measure of developmental stress) were related to several adult cognitive measures. Results are discussed with respect to the best way of measuring avian cognitive abilities, and the utility of developmental telomere attrition as a predictor of adult outcomes.
Highlights
We found weak evidence that faster early growth was associated with slower speed to acquire the reversal
We found some evidence for the importance of early growth for impulsivity, with birds that grew relatively poorly tending to be more impulsive as adults
It is hard to adjudicate between these different interpretations for the effects we found, and the best interpretation may be different for different traits
Summary
Subjects were 37 European starlings (14 female, 23 male), hatched in the wild in May 2013, and taken into captivity before fledging. Despite incorporating a correction for regression to the mean, D was significantly negatively correlated with day 3 T/S ratio (r32 1⁄4 À0.42, P 1⁄4 0.01). This is a commonly observed pattern (Verhulst et al, 2013). The growth rate measure was positively correlated with absolute weight at day 12 (r32 1⁄4 0.70, P < 0.01), and was significantly related to adult skeletal size as indicated by tarsus length at day 24 (r32 1⁄4 0.37, P 1⁄4 0.03). Adult body condition for the study was measured on each entry into the experimental cages, and calculated using the residual from the best-fitting equation relating day 24 weight to skeletal size (as estimated by day 24 tarsus length) for this cohort of birds. Birds were run in replicates of eight consisting of two natal families
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call