The Neural Underpinnings of Associative Learning in Health and Psychosis: How
Can Performance Be Preserved When Brain Responses Are Abnormal?
Graham K. Murray*,1–3, Philip R. Corlett1–4, and Paul C. Fletcher1–3
UK;4Department of Psychiatry, Yale University, New Haven, CT
*To whom correspondence should be addressed; Department of Psychiatry, Box 189, Addenbrooke’s Hospital, University of Cambridge,
Box 189, Cambridge CB2 0QQ, UK; tel: þ44-0-1223-764678, fax: þ44-0-1223-764675, e-mail: firstname.lastname@example.org.
Associative learning experiments in schizophrenia and other
psychoses reveal subtle abnormalities in patients’ brain
responses. These are sometimes accompanied by intact
task performance. An important question arises: How can
learning occur if the brain system is not functioning nor-
mally? Here, we examine a series of possible explanations
for this apparent discrepancy: (1) standard brain activation
patterns may be present in psychosis but partially obscured
by greater noise, (2) brain signals may be more sensitive to
real group differences than behavioral measures, and (3)
patients may achieve comparable levels of performance to
control subjects by employing alternative or compensatory
neural strategies. We consider these explanations in relation
to data from causal- and reward-learning imaging experi-
ments in first-episode psychosis patients. The findings
suggest that a combination of these factors may resolve
the question of why performance is sometimes preserved
when brain patterns are disrupted.
Key words: reinforcement/causal/dopamine/striatum/
uli, thoughts, and percepts is increasingly recognized as
1950s and 1960s produced mixed results,5and for many
years this topic fell out of favor. The majority of research
into cognition in schizophrenia has concentrated instead
on assessing executive control, declarative memory, atten-
a resurgence of interest in the importance of associative
learning in schizophrenia research. This revival may relate
to developments in our understanding of the role of dopa-
domain for survival and environmental adaptation across
ideas have yet to be fully developed, some patterns are
emerging. One notable point is that, in psychosis, there
may be both strengths and weaknesses in learning. This
we selectively review studies on this topic by our own and
other groups and present new analyses of previously pub-
lished imaging studies in which we consider relationships
between brain and behavior during learning.
The Utility of Associative Learning
The British empiricist philosopher David Hume asserted
that all our reasoning is based on associations that we
form.15Associative learning may underpin our ability
to represent and recall the causal and predictive structure
of our environment. It thus provides us with a powerful
means of predicting and therefore perhaps manipulating
the impact of our environment upon us.16
Experiments that present human volunteers with expo-
sure to contingencies between causes and effects indicate
that it is possible to engender the same learning phenom-
ena and biases in human responding upon which formal
animal learning theories are predicated. The same cogni-
tive and neural processes that govern learning about ap-
petitive and aversive events (as studied in experimental
animals or indeed humans) also contribute to human
causal learning and belief formation.7,17,18Key among
these processes is prediction error, the mismatch between
expected and actual experience. Prediction error is sig-
naled by the mesocorticolimbic dopamine and glutamate
Schizophrenia Bulletin vol. 36 no. 3 pp. 465–471, 2010
Advance Access publication on February 12, 2010
? The Authors 2010. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center.
by-nc/2.5), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
systems6,19and contributes to learning about rewards
tion.21Theories that appeal to these processes in the gen-
eration of psychotic symptoms suggest that the neural
and cognitive mechanisms that engage new learning
are deployed erroneously. This would lead to spurious
learning about internal and external events and thereby,
ultimately, to delusional beliefs and hallucinations.
Behavioral Studies of Associative Learning in
Schizophrenia and Other Psychoses Show Specific
Patterns of Strengths and Weakness
When considered as a group, people with psychosis show
an intriguing pattern of strengths and weaknesses in as-
sociative learning.13,22–27While patients are generally im-
paired, this is not always manifest as a slowed or
weakened learning. Rather, there are instances in which
patients can learn things faster than nonpsychotic indi-
viduals. This can occur in case subjects when prior expe-
rience would retard learning in healthy individuals but
fails to do so in patients. For example, in latent inhibi-
tion,28when a healthy individual is preexposed to a stim-
of learning when the preexposed stimulus subsequently
has important consequences. The effect is thought to re-
flect an adaptive mechanism through which attention is
deflected away from redundant environmental cues and
toward more informative stimuli. The attenuation of
latent inhibition in patients with psychosis means that
they learn faster29(though the literature on this is not
consistent30). A consequence of this rapid acquisition
is that patients may attend to and learn about irrelevant
During other feedback-dependent learning tasks,
patients learn slightly more slowly (on average) than con-
trol subjects. However, many schizophrenia patients can
form and maintain the appropriate associations in order
to meet learning criteria for behavioral tasks. For exam-
ple, a recent article documented performance on the
intradimensional/extradimensional (IDED) test from
the Cambridge Neuropsychological Test Automated
Battery on 262 patients from the West London first-ep-
isode psychosis study (232 with schizophrenia) and 76
control subjects.32Although many studies using this
test focus on aspects of attentional set shifting that char-
acterize its latter stages, in fact the task involves partic-
ipants learning from feedback while passing through
various stages of learning. Notably, it has stages of initial
simple discrimination learning, several stages of reversal
learning (where a previously unreinforced stimulus
becomes reinforced), rule abstraction (intradimensional
shifting), and shifting attention to a previously irrelevant
patients showed relatively preserved performance on the
simple discrimination test (though they did, on average,
make significantly more errors than control subjects),
there was a more prominent deficit in reversal learning,
from negative feedback.
These results are consistent with our own study of 119
first-episode psychosis patients using the IDED test,
where we found that psychosis patients made few errors
(though statistically more than control subjects) on sim-
ple reinforcement (simple discrimination learning) in ad-
dition to making more reversal errors.22Other studies
patients made only a few errors, preliminary reports sug-
by strikingly different patterns of brain activity when
compared with control subjects.24,26,35We will now dis-
cuss further how (at the brain level) patients with psycho-
sis are able to learn simple associations and what limits
there are on the success of this learning.
Neuroimaging Studies of Associative Learning Show
Markedly Different Brain Activation Patterns in
Schizophrenia and Other Psychoses Compared With
In 2 previous experiments in (mainly) the same set of
first-episode psychosis case subjects and control sub-
jects, we examined associative learning in the functional
magnetic resonance imaging (fMRI) scanner.24,26One
study involved casual learning, the other probabilistic
ies produced convergent results. Importantly, all the
patients had active psychotic symptoms around the
time of the experiment. Some were taking antipsychotic
medication. In both forms of learning, there was no sig-
nificant difference inlearningrates between patients and
control subjects, but there were remarkable differences
in brain activation between groups. Both studies
revealed learning driven by a mesocorticolimbic net-
work in control subjects but showed that in patients
there was an absence of the normal neural distinction
between important and unimportant events; there was
even some evidence in reward learning that in patients
there was a reversal of the normal pattern of brain ac-
tivation in part of the midbrain. These results are con-
sistent with theories that posit a role for aberrant
incentive salience or dysfunctional prediction error
learning in the pathogenesis of psychotic symptoms.
Furthermore, the greater the magnitude of this dysfunc-
tion (during causal learning), the more severe the delu-
sional ideation of the patient. The findings were not
secondary to antipsychotic medication, as when the
analysis was restricted to a small sample of patients
who were not taking medication, the abnormalities
remained. Twelve months after presentation, the
majority of the patients had received diagnoses of
G. K. Murray et al.
A Paradox: Patients Show Abnormal Brain Responses but
May Learn Normally
We have seen that a proportion of patients with schizo-
tests of associative learning and that this form of learning
may relate to psychotic symptoms. Moreover, the neural
circuitry underlying associative learning appears to be
substantially altered in psychosis. These points prompt
iology of learning is so different in psychosis, how do
patients learn successfully?
There are anumber ofpossibleanswers tothis question.
Perhaps, the standard patterns of activation that underpin
oftherecorded neuralresponsesislower, leadingtosignif-
threshold in patients to see whether a ‘‘normal’’ pattern of
activation is present at a lower grade.
A second possible explanation is that overt behavioral
responses (which quantify learning) are a cruder measure
than that provided by imaging, which is multidimen-
sional. Under more demanding conditions (outside the
scanner), patients may begin to fail but, using reasonably
simple tasks in the scanner, sensitive neural measures are
able to show group differences not detectable in behav-
ioral outcome variables.
A third explanation is that patients may be engaging al-
ternative neural systems to achieve a level of performance
comparable to that of control subjects. Such a possibility
mation available with functional neuroimaging. This
allows us to look beyond the regions of interest and deter-
mine whether patients are engaging brain regions beyond
the normal circuitry. That is, we ignore responses in the
traditional mesocorticolimbic circuitry, which we know
is engaged during prediction error–driven causal learning
and reward learning, and focus instead on activity outside
of these circuits of interest. Care must be taken here, how-
ever. This analysis will not only reveal regions whose ac-
tivity is compensatory but also regions whose responses
may be causing the learning dysfunction, ie, brain areas
whose activation interferes with and is deleterious for suc-
cessful learning. These possibilities can be explored by re-
lating responses in these regions to learning competence.
Below, we consider each of these 3 proposed explana-
tions in more detail, using reexamination of imaging data
during reward learning24and causal learning26to estab-
lish evidence supporting them.
Learning, Albeit to a Lesser Degree Than Control
We reexamined our data and first found support for the
signal-to-noise ratio hypothesis. For example, at a more
lenient statistical threshold than we used previously, we
tum and medial prefrontal cortex during reward learning—
suggesting that patients were activating these areas, just not
threshold, indicative of less robust activation. Perhaps this
modest activity was sufficient for our fairly simple task.
An analogous result was found in the dorsal striatum
by Weickert et al35using an implicit associative learning
task and their performance did not significantly differ
from control subjects, nevertheless, in patients, the dor-
sal, associative striatum, was significantly less active than
in control subjects.
How can patients with reduced or noisy activity
responses show preserved learning? One might ask:
Does the degree of brain activation matter for behavior?
Initial evidence from studies of healthy humans suggests
that reinforcement learning prediction error signal
strength in the dorsal striatum does indeed distinguish
Fig. 1. Reward Prediction Error in Psychosis Patients, Revealing
Activation in Bilateral Ventral Striatum and Medial Prefrontal
Cortex (P < .005 Uncorrected, Minimum Cluster Size 10 Voxels).
Fig. 2. During Causal Learning, Better Performing Patients More
Strongly Activate the A Priori Network of Interest of Head of
Caudate and Right Prefrontal Cortex Compared With Worse
Performing Patients (P < .005 Uncorrected). See online
supplementary material for a color version of this figure.
Associative Learning and Brain Response in Psychosis
better from worse learners during probabilistic learning.
For example, Schonberg et al36examined reward-based
decision making in a sample of more than 30 healthy con-
trol subjects and considered the relationship between
learning performance and brain activation. They found
that striatal prediction error signals during learning differ-
entiated learners from nonlearners and that, across sub-
jects, the magnitude of prediction error signals in the
dorsal striatum correlated significantly with behavioral
performance. We further explored the role of ‘‘signal’’
patients who learned well with responses in those subjects
who were poor learners. We noted that the better learning
patients engaged frontal cortex and dorsal striatum (head
of caudate) to a greater extent than did poorer learners
(figure 2). In this respect, a closer look at the data suggests
that patients do show measurable activations and that
these activations, being smaller, may be sufficient only
to sustain weaker levels of behavioral performance.
Are Brain Signals More Sensitive to Group Differences
Than Behavioral Learning Measures?
There is some evidence in favor of this explanation. In the
causal learningtask,the nonsignificantdifferenceinlearn-
ing measures between patients and control subjects (figure
3) is consistent with the notion that fMRI measures may
indeed represent a more sensitive multidimensional assay
of learning. Indeed, in other work, we have exploited this
sensitivity to adjudicate between competing mechanistic
accounts of causal learning.17In our reward learning
task,24on closer inspection, there was a trend for control
subjectstolearnthereward taskbetter thanpatients (with
a mean of 80% correct choices as opposed to 67% in
patients), but this difference was not statistically signifi-
cant (figure 4, left panel). Furthermore, if reaction times
(as opposed to choice behavior) were used as an index of
learning, there were indeed statistically significant differ-
ences between case subjects and control subjects (figure 4,
right panel). Thus, when viewed across both experiments,
behavioral results in these tasks are less sensitive than im-
aging results at differentiating diagnostic groups, butthey
do reveal some evidence suggestive of group differences.
Fig. 3. Equivalent Learning Rates in First-Episode Psychosis
Patients and Control Subjects During Causal Learning. Error bars
difference was not statistically significant. Right panel shows an adaptive reinforcement-related speeding effect in control subjects
(faster responses on reward trials) and a significant attenuation of this effect in patients. Moreover, patients were significantly faster
than control subjects on the irrelevant, neutral condition. Error bars represent 95% confidence intervals.
G. K. Murray et al.
Do Patients Learn Using Alternative Strategies?
Learning in patients may not be so much impaired as
different, at least in a proportion of patients. Here, neu-
roimaging offers the unique opportunity of building up
an overall picture of not just how patients fail in the task
but how they succeed. It is possible, eg, that patients
may engage additional or alternative neural strategies
in order to achieve behavioral success. If we look at
those patients who are successful (ie, perhaps more
likely to be applying compensatory or alternative mech-
anisms successfully), we can tell whether good perfor-
mance in patients is upheld by differing neural
systems to those responsible for good performance in
control subjects. If patients engage in additional/alter-
native neural activity, and if this extra activity is asso-
ciated with preserved performance, then we could infer
that the extra neural activation represents a strategic or
compensatory change. Thus, as above, this might ex-
plain why patients show (relative) failure of activation
in the traditional neural system in the face of apparently
We examined whether the high-performing patients
(defined usinga mediansplit) specificallyactivated extra
regions making the assumption that a criterion for iden-
tifying such compensatory activity would be that it
would involve regions that were active neither in the
control subjects (where compensatory activity was not
required) nor in the low-performing patients (where fail-
ing performance suggests that compensatory activity is
not occurring or is less prevalent). Patients who were
better learners did not differ from worse learners in
symptomatology but did have higher estimated premor-
bid IQ. In our causal learning data,26we identified
regions (outside of our a priori circuit of interest) that
were more active in competent learning patients (com-
pared with their poor learning groupmates) and further-
more that were not engaged preferentially by better
learning compared with worse learning control subjects.
This combination of contrasts enables us to rule out ac-
tivity that is generically related to better performance,
identifying areas specifically related to performance
boost underpinned by an ‘‘extra’’ or compensatory ac-
tivation. We identified significant foci in the parietal
lobe and the anterior cingulate cortex (see figure 5, up-
per and middle panels). We found that during reward
learning,24the same analytical approach revealed clus-
ters in visual cortex, frontal pole, and right parietal lobe
(figure 5, lower panel).
Given the relative paucity of experimental studies of as-
sociative learning in schizophrenia and taking into ac-
count the increasing theoretical importance of this field,
we believe that considerable further work is required.
Schizophrenia patients demonstrate surprisingly intact
learning in some aspects of associative learning and im-
paired learning elsewhere (especially in cognitive flexi-
bility), but patterns of abnormal brain activity often
emerge in response to tasks where behavioral perfor-
mance does not differ from control subjects. We have
put forward 3 explanations for how patients may ap-
pear to perform well in some learning tasks in spite
of aberrant brain activation. Reexamining data more
closely, we found partial experimental support for
the 3 suggested possibilities. First, patients do show
some normative brain activation at a lower threshold
than control subjects, indicating a partially intact neu-
ral learning system. This is unsurprising as associative
learning is so crucial and so basic aform of learningthat
a completely broken system may be almost incompati-
ble with survival. Second, there is some evidence that
during simple forms of learning, the brain provides
a more sensitive index of learning than behavior
does, with the behavioral abnormality elucidated
only at a more relaxed statistical threshold. Finally,
there is also some evidence that successful learning in
patients is upheld by compensatory mechanisms, ie,
they may engage different strategies from control sub-
jects. Thus, successful learning in patients in reward and
causal learning paradigms may be driven by activation
Strongly Than Worse Performing Patients, Masking Out Areas
Activated by Control Subjects, So Indicating Patient Specificity.
Upper and middle panels show causal learning; lower panels show
reward learning (P < .005). The contrast identified regions that
differentiate better from worse performing patients more than they
patients and masked exclusively by better learning control subjects
greater than worse learning control subjects. See online
supplementary material for a color version of this figure.
Associative Learning and Brain Response in Psychosis
in the frontal pole, anterior cingulate cortex, visual cor-
tex, and parietal cortex. Ultimately, close examinations
of the relationship between brain response and learning
performance are likely to provide richer insights to psy-
chosis than are provided by exploring either measure in
Medical Research Council (Clinician Scientist Award
G0701911 to G. K. M.); NARSAD (Young Investigator
award to G. K. M.); NARSAD (Young Investigator
award to P. R. C.); Medical Research Council and
tute, G0001354); University of Cambridge (Parke Davis
Exchange Fellowship to P. R. C.); Wellcome Trust
(Senior Research Fellowship in Clinical Science award
064351 to P. C. F.); Bernard Wolfe Professorial award
in Health Neuroscience to P. C. F.
1. Pavlov IP. Conditioned Reflexes: An Investigation of the Phys-
iological Activity of the Cerebral Cortex. New York, NY:
Dover Publications Inc; 1927.
2. Locke J. An Essay Concerning Human Understanding.
London: Printed by E. Holt, For T. Basset; 1690.
3. Bleuler E. Dementia Praecox or the Group of Schizophrenias.
New York, NY: International University Press; 1911/1950.
4. Schneider K. Clinical Psychopathology. New York, NY:
Grune and Stratton; 1959.
5. Buss AH, Lang PJ. Psychological deficit in schizophrenia: I.
Affect, reinforcement, and concept attainment. J Abnorm
6. Schultz W, Dickinson A. Neuronal coding of prediction
errors. Annu Rev Neurosci. 2000;23:473–500.
7. Dickinson A. The 28th Bartlett Memorial Lecture. Causal
learning: an associative analysis. Q J Exp Psychol B.
8. Gray JA, Feldon J, Rawlins JNP, Smith AD. The neuropsy-
chology of schizophrenia. Behav Brain Sci. 1991;14:1–19.
9. Miller R. Schizophrenic psychology, associative learning and
the role of forebrain dopamine. Med Hypotheses. 1976;
10. Corlett PR, Honey GD, Fletcher PC. From prediction error
to psychosis: ketamine as a pharmacological model of delu-
sions. J Psychopharmacol. 2007;21:238–252.
11. Kapur S. Psychosis as a state of aberrant salience: a frame-
work linking biology, phenomenology, and pharmacology
in schizophrenia. Am J Psychiatry. 2003;160:13–23.
12. Heinz A, Schmidt LG, Reischies FM. Anhedonia in schizo-
correlates. Pharmacopsychiatry. 1994;27(suppl 1):7–10.
13. Murray GK, Fletcher PC. Can models of reinforcement
learning help us to understand symtpoms of schizophrenia?
In: Dreher J-C, Tremblay L, eds. Handbook of Reward and
Decision-Making. Oxford, UK: Elsevier; 2009:251–269.
14. Murray G. Dopamine dysfunction and delusions, hallucina-
tions and anhedonia. Eur Psychiatr Rev. 2009;2:21–23.
15. Hume D. An Enquiry Concerning Human Understanding.
London: A. Millar; 1748.
16. Dickinson A. Conditioning and associative learning. Br Med
17. Corlett PR, Aitken MR, Dickinson A, et al. Prediction error
during retrospective revaluation of causal associations in
humans: fMRI evidence in favor of an associative model of
learning. Neuron. 2004;44:877–888.
18. Aitken MR, Larkin MJ, Dickinson A. Super-learning of
causal judgements. Q J Exp Psychol B. 2000;53:59–81.
19. Lavin A, Nogueira L, Lapish CC, Wightman RM, Phillips
PE, Seamans JK. Mesocortical dopamine neurons operate
in distinct temporal domains using multimodal signaling.
J Neurosci. 2005;25:5013–5023.
20. Matsumoto M, Hikosaka O. Two types of dopamine neuron
distinctly convey positive and negative motivational signals.
21. Bromberg-Martin ES, Hikosaka O. Midbrain dopamine neu-
rewards. Neuron. 2009;63:119–126.
22. Murray GK, Cheng F, Clark L, et al. Reinforcement and re-
versal learning in first-episode psychosis. Schizophr Bull. 2008;
23. Murray GK, Clark L, Corlett PR, et al. Incentive motivation in
first-episode psychosis: a behavioural study. BMC Psychiatry.
tegmental reward prediction error disruption in psychosis. Mol
25. Gray JA. Integrating schizophrenia. Schizophr Bull. 1998;24:
26. Corlett PR, Murray GK, Honey GD, et al. Disrupted
prediction-error signal in psychosis: evidence for an asso-
ciative account of delusions. Brain. 2007;130(pt 9):2387–
27. Waltz JA, Frank MJ, Robinson BM, Gold JM. Selective re-
inforcement learning deficits in schizophrenia support predic-
dysfunction. Biol Psychiatry. 2007;62:756–764.
28. Lubow RE. Latent inhibition: effects of frequency of nonrein-
forced preexposure of the CS. J Comp Physiol Psychol. 1965;
29. Martins Serra A, Jones SH, Toone B, Gray JA. Impaired as-
sociative learning in chronic schizophrenics and their first-
degree relatives: a study of latent inhibition and the Kamin
blocking effect. Schizophr Res. 2001;48:273–289.
30. Rascle C, Mazas O, Vaiva G, et al. Clinical features of latent
inhibition in schizophrenia. Schizophr Res. 2001;51:149–161.
31. Lubow RE, Kaplan O. The visual search analogue of latent
inhibition: implications for theories of irrelevant stimulus
processing in normal and schizophrenic groups. Psychon
Bull Rev. 2005;12:224–243.
32. Leeson VC, Robbins TW, Matheson E, et al. Discrimination
learning, reversal, and set-shifting in first-episode schizophre-
nia: stability over six years and specific associations with med-
ication type and disorganization syndrome. Biol Psychiatry.
models of striatal-cortical
G. K. Murray et al.
33. McKirdy J, Sussmann JE, Hall J, Lawrie SM, Johnstone EC,
McIntosh AM. Set shifting and reversal learning in patients
with bipolar disorder or schizophrenia. Psychol Med.
34. Waltz JA, Gold JM. Probabilistic reversal learning impair-
ments in schizophrenia: further evidence of orbitofrontal dys-
function. Schizophr Res. 2007;93:296–303.
35. Weickert TW, Goldberg TE, Callicott JH, et al. Neural corre-
lates of probabilistic category learning in patients with schizo-
phrenia. J Neurosci. 2009;29:1244–1254.
36. Schonberg T, Daw ND, Joel D, O’Doherty JP. Reinforce-
ment learning signals in the human striatum distinguish learn-
ers from nonlearners during reward-based decision making.
J Neurosci. 2007;27:12860–12867.
Associative Learning and Brain Response in Psychosis