Auditory stimulus discrimination recorded in dogs, as indicated by mismatch negativity (MMN)

Dog cognition research tends to rely on behavioural response, which can be confounded by obedience or motivation, as the primary means of indexing dog cognitive abilities. A physiological method of measuring dog cognitive processing would be instructive and could complement behavioural response. Electroencephalogram (EEG) has been used in humans to study stimulus processing, which results in waveforms called event-related potentials (ERPs). One ERP component, mismatch negativity (MMN), is a negative deflection approximately 160–200 ms after stimulus onset, which may be related to change detection from echoic sensory memory. We adapted a minimally invasive technique to record MMN in dogs. Dogs were exposed to an auditory oddball paradigm in which deviant tones (10% probability) were pseudo-randomly interspersed throughout an 8 min sequence of standard tones (90% probability). A significant difference in MMN ERP amplitude was observed after the deviant tone in comparison to the standard tone, t5 = −2.98, p = 0.03. This difference, attributed to discrimination of an unexpected stimulus in a series of expected stimuli, was not observed when both tones occurred 50% of the time, t1 = −0.82, p > 0.05. Dogs showed no evidence of pain or distress at any point. We believe this is the first illustration of MMN in a group of dogs and anticipate that this technique may provide valuable insights in cognitive tasks such as object discrimination.

► Dog behaviour may be affected by motivation, learning experience, training levels.
► A physiological measure of cognitive processing could complement behaviour research.
► Mismatch negativity (MMN) is related to stimulus discrimination in humans.
► MMN research in animals can be invasive, but there are less invasive techniques.
► We present data showing MMN in dogs during an auditory oddball paradigm.