Introduction: Many emotional processes are affected by sleep restriction (Beattie et al. 2015). Whether this is likewise true for social emotions, such as empathy, is not known. Empathy for pain has previously been studied using paradigms where subjects are presented with pain in others or pictures of pain in others. These paradigms consistently activated areas in the anterior cingulate cortex and anterior insula. Aging affects both sleep (Vitiello 2012) and emotional functions (Mather 2012), but whether the role of sleep in emotional functioning is stable across age is not known. This study aims to investigate how neural and behavioral responses to pain in others are affected by sleep restriction and age, and whether age modulates the role of sleep in responses to pain in others.

Methods: In a randomized cross-over experimental design, 47 healthy younger (age: 20-30) and 39 older (age: 65-75) volunteers underwent fMRI twice, after either normal sleep or sleep restricted to 3 hours. In an event-related fMRI task, participants viewed pictures of needles pricking a hand (pain condition) or Q-tips touching a hand (control condition), and reported their vicarious unpleasantness. Preprocessing and analyses were performed in SPM12 and included slice time correction, realignment, DARTEL normalization and smoothing with an 8x8x8 FWHM kernel. First level analyses included fixed effects for events, motion parameters and button presses. At second level a full factorial design was applied. Additional region of interest analyses were performed in anterior insula and anterior cingulate cortex.

Results: The contrast pain > control robustly activated anterior cingulate cortex and anterior insula (FWE p < 0.05, fig 1) as well as other areas previously proposed as the core empathy for pain network (Lamm et al. 2011). Older participants generally experienced more unpleasantness in response to pictures of pain compared to younger participants (p < 0.001), and this was accompanied by higher activity in bilateral angular gyrus (FWE p < 0.05). Age and sleep interacted so that sleep restriction caused decreased unpleasantness in young and increased unpleasantness in old to pain stimuli (p < 0.01), even though there was no significant simple main effect of sleep restriction in any age group. In clusters in bilateral insula, old participants showed more activity and young less activity in response to pain after sleep restriction (p < 0.001 uncorrected).

Conclusions: Compared to younger participants, older subjects generally responded more to pain in others, shown as subjective experience as well as brain responses. With sleep restriction, empathic responses in young and old changed in opposite directions, so that empathic responses increased in older and decreased in younger participants. Given that empathy is crucial in effective interaction with others, our findings imply possible age-related changes in prosocial behavior, amplified by short sleep.