Encoding information into memory is sensitive to distraction whilst retrieving that memory may be compromised by proactive interference from pre-existing memories. These two debilitating effects are common in neuropsychiatric conditions but modelling them preclinically to date is slow as it requires prolonged operant training. A step change would be the validation of functionally equivalent but fast, simple, high-throughput tasks based on spontaneous behaviour. Here, we show that spontaneous object preference testing meets these requirements in the subchronic phencyclidine (scPCP) rat model for cognitive impairments associated with schizophrenia. scPCP rats show clear memory sensitivity to distraction in the standard novel object recognition task (stNOR). However, due to this, stNOR cannot assess proactive interference. Therefore, we compared scPCP performance in stNOR to that using the continuous NOR task (conNOR), which offers minimal distraction, allowing disease-relevant memory deficits to be assessed directly. We first determined that scPCP treatment did not affect whisker movements during object exploration. scPCP rats exhibited the expected distraction stNOR effect but had intact performance on the first conNOR trial, effectively dissociating distraction by using two NOR task variants. In remaining conNOR trials, scPCP rats performed above chance throughout but, importantly, their detection of object novelty was increasingly impaired relative to controls. We attribute this effect to the accumulation of proactive interference. This is the first demonstration that increased sensitivity to distraction and proactive interference, both key cognitive impairments in schizophrenia, can be dissociated in the scPCP rat using two variants of the same fast, simple, spontaneous object memory paradigm.