Animals living in temperate regions prepare for harsh winter conditions by responding to environmental cues that signal resource availability (e.g., food, day length). Siberian hamsters (Phodopus sungorus) breed in long, summer-like days (LD, >14 h light), i.e., photoperiods, and undergo robust gonadal regression and become more aggressive when exposed to short, winter-like photoperiods that signal impending limited resources (SD, <10 h light). When hamsters are reared within an intermediate photoperiod (ID, 13.5 h light), they are re-productively active, but undergo gonadal regression in response to mild food restriction (FR) over 6–12 weeks. We hypothesized that short-term (1–2 weeks) FR inanID photoperiod would provide a signal of impending limited resources and initiate the seasonal increase in aggression typical of SD photoperiods, as well as alter reproductive behaviors in advance of gonadal regression. To test this, we housed male and female hamsters in LD or ID photoperiods, with ad libitum (AL) access to food or a 90%-AL ration. We tested aggressive behavior after one week and reproductive behavior after two weeks, and subsequently monitored females for pregnancy and litter production. Both sexes displayed increased aggression in the ID-FR treatment. Untreated male intruders were less likely to ejaculate when paired with ID females during reproductive encounters. ID-FR males were undergoing gonadal regression after two weeks, but were more likely to have ejaculated. Female pregnancy and litter characteristics were unaltered by treatment: females were equally likely to achieve pregnancy and produce comparable litters across treatment groups. Collectively, we demonstrate that a signal of diminishing resources in an ID photoperiod is sufficient to trigger seasonal aggression, but that hamsters are reproductively resilient to inhibitory environmental cues in the short term. Broadly, our findings provide an important context for exploring seasonal changes in behavior and physiology from an ultimate perspective.

This cross-sectional measurement study demonstrates a technique for combining information from several aggression scales into one aggression score using latent variable modeling. Participants included male patients (n = 49) with a DSM-IV diagnosis of dementia at The Veterans Affairs Medical Center Outpatient Geriatric Psychiatry Clinic. Data from seven aggression scales were used to indicate the latent aggression variable. Results provided evidence that a unidimensional latent variable model of aggression adequately represented the data. Reliability of the aggression latent variable was estimated as 0.90, whereas reliability of the separate scales estimated with this sample were less than 0.84. Our findings suggest that combining multiple scales into one aggression score using latent variable modeling results in comprehensive and reliable aggression scores that offer researchers several advantages over current methods for measuring aggression.