OADES, R.D. (1997). Stimulus dimension shifts in patients with schizophrenia, with and without paranoid hallucinatory symptoms, or obsessive compulsive disorder: strategies, blocking and monoamine status. Behavioural Brain Research, 88, 115-131. (request a copy)

In accord with our understanding of journal policy, we present the pre-publication text (view).
The final version (cited above: doi:10.1016/S0166-4328(97)02304-8 ) is available at http://www.sciencedirect.com/science?_ob=PublicationURL&_tockey

Introduction: Reversal and intra-dimensional (ID) and extra-dimensional (ED) non-reversal shifts in task discrimination learning were compared. The aim was to see if "learned inattention" to the irrelevant dimension differentially influenced the efficacy of learning and of the stimulus choice strategy. (An overall indicator of monoamine metabolism was measured for potential congruence between switches of attention and dopamine activity: see Oades, Neurosci. Biobehav. Rev., 9, 261-283, 1985).

Methods: Performance on pattern-discrimination discrimination shifts was compared with conditioned blocking (CB: another test of "learned inattention") and related to the status of monoamine neurotransmitter metabolism reflected in 24h-urine samples between tests. Results are reported for 29 healthy subjects (mean age 18.0y), 13 patients with obsessive compulsive disorder (OCD: mean age16.3y), and 28 with schizophrenia, including 14 paranoid hallucinatory (PH: 19.6y) and 14 nonparanoid patients (NP: 17.5y).

1. PH and NP patients improved learning with practice but showed an impaired shift on each task.
2. Unlike PH and control subjects, the NP shift impairment was non-specific and related to their problems on simple reversal of the discrimination.
3. The length of the stimulus-response sequences showed that all subjects were able to acquire a set for colour.
Figure 14. An analysis of choice on sequential pairs of stimuli showed that while all patients showed fewer win-stay sequences, only PH patients perseverated with lose-stay sequences. This type of error in PH patients contrasts with the increase of win-shift errors in NP patients (figure 1).
5. Learning about the added stimulus on the CB task related to the efficiency of intra-dimensional shift in NP patients.
6. An impairment with OCD patients was restricted to the ED-shift -- (not reversal or ID-shift).
7. Increases of dopamine activity related to slower initial learning, but to more switches (and rapid learning) on all shift tasks: (positive correlations with win/lose-shift, negative with win-stay). NA activity in PH and NP patients related to increased win-stay and decreased lose-shift decisions (figure 1).
Figure 28. Increased serotonin activity correlated with faster learning in controls, OCD and PH patients. But the opposite relationships for dopamine and serotonin activity held for NP patients (figure 2).

The different tasks of the "learned inattention" paradigm have different if related requirements and correlates. The monoamine data are consistent with the postulated function of noradrenaline in tuning and dopamine in switching operations. The behavioural data are consistent with the automatization of endogenous information processing, while NP patients use exogenous attentional strategies for selecting information and PH patients show inefficient endogenous control of attention.