Impact of cognitive symptoms and anhedonia on functioning in depression

Department of Psychiatry
University of Oxford
Warneford Hospital
Oxford, UK

Impact of cognitive symptoms and anhedonia on functioning in depression

by G. M. Goodwin, United Kingdom

It is convenient to distinguish the cognition associated with the emotions, sometimes called hot cognition, from the cold cognition that is associated with simple acts of remembering, calculating, planning, or executing complicated but essentially emotion-free tasks. While it is unsurprising that hot cognition is distorted by extreme mood states like major depression and mania, it would not necessarily follow that cold cognition be also affected. In fact, the centrality of impaired cold cognition has now been empirically established for major depression, together with the related problem of psychomotor slowing. Nevertheless, actual measurement using standard methods from neuropsychology is not a standard of care in psychiatry. It probably should be, because persisting symptoms of impaired episodic memory and reduced motivation are probably linked to the subjective experience of anhedonia. Moreover, they limit the capacity for full functional recovery from major depression. Experimental studies in animals and man suggest that the cognitive impairments of depression are at least partially reversible by antidepressants. This implies that improved outcomes in major depression may be achievable by targeting cognition rather than just depressive symptoms. Hot cognition has been the obvious focus for understanding mood disorder, and cognitive behavior therapy was developed as a specific remedy for the extreme negative cognitive distortions of the depressed state. Its efficacy is usually taken to be a reasonable proof that the cognitive mechanisms it targets are actually a fundamental part of the illness. It poses a secondary question, which is how antidepressants might work on essentially the same processes. This can now be studied more easily because neuroscience has taught us that there is a bridge between hot and cold cognition via studies of motivation and reward. This is encouraging a more subtle approach to distinguishing different components of emotion, particularly anhedonia, and the contribution of pleasure to full functional recovery from depression.

Medicographia. 2014;36:446-451 (see French abstract on page 451)

Cognition is usually defined as the mental process whereby knowledge and understanding is acquired through thought, experience, and the sense organs. The use of the word in psychology has extended to include active processes such as judging and problem solving, imagination, and even planning. Accordingly, it has come to encompass, by such extension, essentially emotional components of experience because choice and decision-making are inherently based on the values that form the basis of our emotional life. It nevertheless remains convenient to distinguish the cognition associated with the emotions, sometimes called hot cognition, from the cold cognition that is associated with simple acts of remembering, calculating, planning, or executing complicated but essentially emotion-free tasks. Like all such distinctions, the hot/cold dichotomy is valid for the extreme examples of the two kinds of cognition but breaks down when one approaches more nuanced mental processes.

Cold cognition is usually taken to include attention, memory, and executive function.1 These are all nouns that have ordinary language meanings and we can all give subjective accounts of how we believe our attention, memory, or executive function to be at any given moment. Accordingly, psychologists and psychiatrists sometimes use entirely subjective scales to decide whether or not individuals have impaired cognition of this kind. Subjective complaints about memory, for example, may precede the onset of frank dementia but they are also clearly a part of major depression. Indeed, if one observes the criteria for a major depressive episode in DSM-5, shown in Table I,2 the 8th is “reduced concentration with diminished ability to think or concentrate, or indecisiveness.” So this kind of cognitive impairment is incorporated in the simplest definitions of a major depressive episode. Indeed, the cognitive impairment that it implies is frequently the reason why people may seek the attention of a doctor. Their inability to conduct daily activities that rely on simple memory, concentration, and decision-making may even define the severity of the depressed state and the requirement for treatment or, in the worse cases, admission to hospital. Admission to hospital may often be required because of the complete failure of a patient’s executive capacity. Clearly, recovery of cold cognition is a necessity for recovery of everyday function.

Table I
Table I. Symptoms of a major depressive episode.

After reference 2: American Psychiatric Association. Diagnostic and Statistical
Manual of Mental Disorders (5th Ed). Arlington, VA: American Psychiatric Publishing;
2013. © 2013, American Psychiatric association.

Despite the centrality of impaired cognition and the related problem of psychomotor slowing in the presentation of depressed patients, actual measurement using standard methods from neuropsychology is not a standard of care in psychiatry This is regrettable because, in fact, measures of memory, attention, and executive function can be made relatively easily both in a laboratory setting but also in a well-organized clinic. The measures that are used are traditionally pencil-and paper-based,3 but are now of course often also available in a form that can be presented via computer screens and keyboards, which can make data curation much easier.

Despite the centrality of impaired cognition and the related problem of psychomotor slowing in the presentation of depressed patients, actual measurement using standard methods from neuropsychology is not a standard of care in psychiatry This is regrettable because, in fact, measures of memory, attention, and executive function can be made relatively easily both in a laboratory setting but also in a well-organized clinic. The measures that are used are traditionally pencil-and paper-based,3 but are now of course often also available in a form that can be presented via computer screens and keyboards, which can make data curation much easier.

Measuring cold cognition

Sustained attention is the ability to remain on target during boring repetitive tasks. It involves keeping in mind perceptions and instructions that allow relatively simple online decisions to be made. For example, if one is read out a sequence of numbers, one can determine how many numbers can be remembered in a sequence, either forwards or backwards. This socalled“ digit span” test is a useful but rough-and-ready measure of attention.3 A somewhat more credible measure is from continuous performance tasks where the individual is required to view a screen with a sequence of symbols or numbers and the instruction is to identify when, for example, a particular sequence of numbers occurs. This can be programmed so the targets are relatively infrequent and the numbers change rapidly, which makes the task more difficult. Such tasks are subjectively effortful and there may be an interesting association with the effort of producing maximal handgrips for example,4 because similar central processes may be involved.

Episodic memory
Episodic memory is usually measured using lists of nouns (typically 16), which are repeated on 5 occasions after each of which the subject is tested for immediate recall. Over the 5 tests the number recalled increases and the performance can also be evaluated after a delay, with intervening distraction. This test gives some idea of the rate at which learning occurs, the learning capacity (in other words the maximum number of words that can be remembered by an individual), and consolidation/recall.3

Executive function
Executive function is the term used to describe a host of different kinds of actions that have in common the prioritization and selection between different alternatives. In its simplest form this ability to switch set can go wrong, as an inability to learn that a particular contingency has changed. So if in a task that chooses between two different buttons, one button is the correct one in the first part of the test and is rewarded but the second button becomes the correct button after a finite period of time and without warning, then normally an adaptive response, when reward is withheld from button 1, is to try the other button and to discover that it is now the correct one. If this change of set fails to occur quickly, it is described as perseveration. Perseverative behavior is very characteristic of patients with major frontal lobe lesions and more subtle disturbances of executive function are often assumed to reflect abnormalities of frontal lobe function, or perhaps more correctly the functional networks that are routed through the frontal cortex.3

Disturbances of cold cognition in major depression

All domains of cognitive function have been described as abnormal in major depression, particularly when patients are acutely depressed.1,5 The pattern of deficits is similar in younger and older groups, although memory and learning are more affected in those over 60.6 In the past, this was often discounted on the basis that patients were not trying, or were simply poorly motivated to cooperate with the experimental procedure. However, there is little evidence to suggest that this is actually the case, so for over a decade now the general consensus has been that the measured abnormalities in neuropsychological function should be accepted as genuine measures of the patient’s ability to perform tasks.1 Moreover, failure on these tasks has important implications for real life functioning that will limit recovery and rehabilitation. There is, therefore, every reason to focus in the future on cold cognition in the design of treatment and management for patients with major depression.

The degree of memory impairment correlates modestly with the severity of depression measured on scales of largely subjective reports like the Hamilton Rating Scale for Depression (or HAM-D) and show a diurnal swing in the course of a day.4 This implies that the underlying neurobiology is linked to depression per se and that it may be quite quickly improved as mood improves. Meta-analysis has illustrated the magnitude and relative consistency of the effects. It confirms that formal tests should concentrate on attention, and particularly, memory.

The disturbance of memory function broadly recovers with treatment, even for example when the treatment is electroconvulsive therapy (ECT), but there have been relatively few clinical trials in which cognition has been measured as a potential treatment target for antidepressants. One unusual example found that a measure of speed/executive function improves significantly during treatment with duloxetine or the new antidepressant vortioxetine in elderly patients.7 It is not established what the best cognitive treatment target would be in major depression but one clue is the persistence of memory impairment following recovery from acute episodes. This has been looked at in a large general practice sample that suggested that the extent of impairment between episodes was related to the number of previous episodes the patient had suffered.8 This is compatible with the hypothesis that repeated episodes of depression are, in some sense, neurotoxic and result in accumulating impairments of cognition, which again may contribute to the difficulties such patients face in returning to work and the attainment of premorbid levels of function. The persistence of cognitive impairments in euthymic patients is nevertheless most striking in older patients and is seen across different domains relatively unselectively.9

In depression, executive function is obviously disturbed because of the inability of the subject’s experience in conducting the business of daily life. In terms of symptoms, as shown previously in Table I, indecisiveness is a key part of item 8. It has two elements, one is a slowing and the other is a failed capacity to make decisions at all. These failures can be captured in the laboratory in a relatively straightforward way by tests that require simple alternation between different targets.

The simplest is the Trail Making Test (TMT), which requires subjects to join the dots between letters or words. Part A of the TMT simply determines speed (and comprehension of the test): it requires following either numbers or letters. The second part requires alternation between the sequence of letters and numbers; it is just about the simplest test of executive function one could devise and its primary output is the speed at which the task is completed.3 However, it would be possible to capture errors where subjects fail to actually make the necessary switch and, for example, continue to join letters rather than alternating to numbers. In depression, the dysfunction is primarily expressed as slowing in both phases of the TMT. Thus, in practice, measures of executive function in depression usually rely on an estimate of the speed to complete relatively simple executive tasks.

Another test that does this quite effectively and has been used in clinical trials in depression is the digit symbol substitution test, which requires subjects to copy a list of 9 symbols that have to be matched with corresponding numerical digits in a random sequence.3 The time taken to complete a fixed number of images or the number of images completed within a fixed time gives the measure. Tests of executive function are usually related to measures of intelligence simply because measures of intelligence are aggregated average measures of tests of executive function. It is therefore often useful to have an independent language-based measure of premorbid intelligence in patients with depression. The National Adult Reading Test is available in English (and other languages) to provide this estimate.10 It relies on the correct pronunciation of irregular words and correlates again with levels of education.

The pharmacology of cognitive impairment in depression

Most of the existing studies that have direct relevance to the use of antidepressants in general, and the selective serotonin reuptake inhibitors (SSRIs) in particular, relate to the role of serotonin in cold cognition. In man, tryptophan depletion has been taken to be one way of modestly reducing serotonergic function in the brain11: different loading doses of competing amino acids can produce different degrees of tryptophan depletion (TD). Across a wide range of studies the most consistent effects of tryptophan depletion was impairment of episodic memory, particularly that requiring verbal learning.12 This effect did not seem to extend to spatial memory or semantic or working memory or, indeed, executive function and sustained attention. These effects on memory may be more striking in individuals with a previous history of depression.13 It is notable that the effects of low level TD are obtained largely independently of effects on mood per se. Frank mood effects require high level TD and were first described in recovered patients receiving SSRI treatment.14 Comparable effects are also seen in recovered patients not on treatment,15 but are only seen in healthy volunteers if they have a family history of mood disorder.16 In fact, in vulnerable subjects there is an impact both on cold cognition (memory) and hot cognition (shown by negative cognitive bias but not frank depressive symptoms in patients at risk of depression).13

The findings using drugs that putatively increase serotonin availability have been mixed. Thus, for example, following acute infusion of citalopram an effect on memory consolidation was found in healthy women, which, in part at least, compliments the corresponding findings from tryptophan depletion.17 Other studies have found a variety of other effects, often of impairment, which are difficult to interpret. There is a much more extensive and evolving data set in animals, which is beyond the scope of this article. Generally, it appears that effects of 5-hydroxytryptamine (5-HT) receptor subtype manipulation on learning and memory are exerted through alterations in the release of neurotransmitters such as acetylcholine and glutamate, which have been more directly implicated in cognitive function than 5-HT itself.18 It suggests the potential for modulation of attention and learning by a range of transmitter systems.

Hot cognition

Hot cognition is the most obviously disturbed and disturbing domain of psychological function for the depressed patient. The primary symptoms of depressed mood are markedly diminished interest or pleasure, feelings of worthlessness or excessive guilt, and recurrent thoughts of death or suicidal ideation. They all represent the powerful emotional content that shapes the experience and presentation of major depression. In addition, the patients may describe these experiences as literally painful, and pain may indeed be the primary complaint of patients who present with depression. Hot cognition may be a literal barrier to normal personal and social function.

Cognitive behavior therapy
The hot cognitions that are so obvious in conscious experience have given rise to at least one very useful way of thinking about the experience of patients with depression. Half a century ago Aaron Beck identified negative thinking as a central feature of all states of depression.19 Such thinking focuses on oneself, the world, and the future. This triad in which suicidal ideation is particularly important struck him as an important maintaining factor in all types of depression. This conscious negativity provides the focus of cognitive behavior therapy. Cognitive behavior therapy takes what is essentially a Socratic approach to the patients’ experience and explores whether their beliefs and convictions about the world itself and the future are actually proportionate. The role of the therapist is to change the perceptions and experiences of the patient by reasoning. In doing so, therapists elicit the extremes of conscious emotional experience described by patients and become involved with the patients in ways that are hard to measure. The essential ingredients of cognitive behavior therapy (CBT) are therefore often confounded by nonspecific factors. For this reason, psychotherapy research faces considerable challenges in its choice of fair comparison treatments, but a relatively large literature supports its efficacy in major depression.20

The efficacy of CBT is usually taken to be a reasonable proof that the cognitive mechanisms it targets are actually a fundamental part of the illness. It poses a secondary question, which is how antidepressants might work on essentially the same processes. It turns out that they also reverse cognitive biases, but of course not simply at a conscious level. Instead, measures of automatic emotional bias relating to the perception of the emotional content of faces or the memory for positive and negative material is both biased in a negative way in patients and reversed in a relatively short term by antidepressants.21 This convergence of antidepressant effects and CBT effects is of great current interest and strongly suggests the complementarity of these traditionally opposed forms of therapy.

The spectrum between hot and cold: anhedonia, warm cognition, motivation
At first sight the processes of hot cognition appear very different from the cold cognition that has already been described. However, while these extremes may appear qualitatively distinguishable at a clinical level, a clear gradient intervenes to join extremes of emotional experience through to emotionally valenced decision and motivation. When we refer to absence of motivation we imply the absence of the stimulus/ incentive to act or react in a certain way. Purposeful behavior is motivated behavior, which means that either physiological or social stimuli activate or motivate a person to do something. The neuroscience of this kind of behavior is a major current interest, quite independent of its implications for studying clinical depression.22 The current understanding can be summarized by reference to Figure 1. The vertical axis represents action versus inhibition; the horizontal axis represents reward versus punishment. Movement across this two-dimensional space will be influenced by the balance between these dimensions, modulated by the environment. What is of great interest for psychiatry is the implication that normal function will require both sensitivity to reward and the necessary vigor to respond, and that these could be independent/uncoupled. Furthermore, our nervous systems may have evolved in environments where rewards were quite infrequent and their pursuit would be quite simple to learn, but punishment or danger would be ever present and require more complex learning. These opposed functional outcomes may depend on the balance between different neurotransmitter systems with reward/ invigoration modulated by dopamine/noradrenaline and punishment/ inhibition by serotonin/noradrenaline. This description derives from work on learned behavior in animals, but maps on to the so-called circumplex theory of affect, which in one form or another, has enjoyed a long prominence in the human psychology literature. This literature puts positive arousal/ passivity orthogonal with a negative dimension of sad/happy mood (Figure 1).23

Figure 1
Figure 1. The affective circumplex.

Dopamine/(noradrenaline) is associated with: prospect of reward or error signal,
vigor/action/approach, and adaptation to a world of rare treats. Serotonin is associated
with: dangers and threats that have to be managed, behavioral inhibition,
context, prediction, and memory.

Clinically, the recognition that both the depressed state and the recovery after depression may involve more than one dimension (of clinical symptoms) remains novel. Instead, absent pleasure and loss of vigor are often assumed to co-vary. The structure of normal emotion instead would predict differential recovery. Anhedonia is probably the psychopathological construct that best reflects the pleasure domain. It means the absence of pleasure from acts that would normally be pleasurable. However, it has received relatively little attention in clinical research despite being prominent as an item in the diagnosis of major depression (item 2 in Table I). The Snaith-Hamilton Pleasure Scale (SHAPS) attempts to capture the pleasure likely to be experienced under a range of conditions (watching TV, being with family or friends, hobbies and pastimes, favorite meal, warm bath or shower, etc).24 It is essentially a judgment of how things have recently been and it does not require actual participation in these activities. Nevertheless, it clearly tests a different experience from simply asking about depressive symptoms and so is quite likely to differentiate from a scale like the HAM-D scale.25 Preliminary trials suggest that this is the case for agomelatine, compared with venlafaxine. Thus, agomelatine showed the same effect as venlafaxine on the HAM-D scale, but a better recovery on the SHAPS scale.26


Cold cognition is an unexpected accompaniment of mood disorder. It is easy to measure and may be amenable to treatment. It is fundamental to our understanding of full recovery of function in major depression. Hot, or even warm, cognition is a construct that also encourages us to look beyond the simple symptoms of a depressed mood. In either case, cognitive function provides the stimulus to take a more sophisticated approach to understanding antidepressant action and the potential differences between antidepressants. The interest of neuroscientists in the structure of emotional experience and learning means that there are relatively simple experiments that will allow the dissection of warm cognition in laboratory settings. Warm cognition can be tested in a more mechanistic way than the simple experience and observation of emotion could allow. It is a first step to more personalized diagnosis and treatment of mood disorder and a more complete restoration of function.

1. Austin MP, Mitchell P, Goodwin GM. Cognitive deficits in depression: possible implications for functional neuropathology. Br J Psychiatry. 2001;178:200-206.
2. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (5th Ed.). Arlington, VA: American Psychiatric Publishing; 2013.
3. Lezak MD. Neuropsychological assessment (5th ed). New York, NY: Oxford University Press; 2012.
4. Moffoot AP, O’Carroll RE, Bennie J, et al. Diurnal variation of mood and neuropsychological function in major depression with melancholia. J Affect Disord. 1994;32:257-269.
5. Austin MP, Ross M, Murray C, O’Carroll RE, Ebmeier KP, Goodwin GM. Cognitive function in major depression. J Affect Disord. 1992;25:21-29.
6. Thomas AJ, Gallagher P, Robinson LJ, et al. A comparison of neurocognitive impairment in younger and older adults with major depression. Psychol Med. 2009;39:725-733.
7. Katona C, Hansen T, Olsen CK. A randomized, double-blind, placebo-controlled, duloxetine-referenced, fixed-dose study comparing the efficacy and safety of Lu AA21004 in elderly patients with major depressive disorder. Int Clin Psychopharmacol. 2012;27:215-223.
8. Gorwood P, Corruble E, Falissard B, Goodwin GM. Toxic effects of depression on brain function: impairment of delayed recall and the cumulative length of depressive disorder in a large sample of depressed outpatients. Am J Psychiatry. 2008;165:731-739.
9. Bora E, Harrison BJ, Yucel M, Pantelis C. Cognitive impairment in euthymic major depressive disorder: a meta-analysis. Psychol Med. 2013;43:2017-2026.
10. Nelson HE, O’Connell A. Dementia: the estimation of premorbid intelligence levels using the New Adult Reading Test. Cortex. 1978;14:234-244.
11. Young SN. Acute tryptophan depletion in humans: a review of theoretical, practical and ethical aspects. J Psychiatry Neurosci. 2013;38:294-305.
12. Cowen P, Sherwood AC. The role of serotonin in cognitive function: evidence from recent studies and implications for understanding depression. J Psychopharmacol. 2013;27:575-583.
13. Hayward G, Goodwin GM, Cowen PJ, Harmer CJ. Low-dose tryptophan depletion in recovered depressed patients induces changes in cognitive processing without depressive symptoms. Biol Psychiatry. 2005;57:517-524.
14. Delgado PL, Charney DS, Price LH, Aghajanian GK, Landis H, Heninger GR. Serotonin function and the mechanism of antidepressant action: Reversal of antidepressant- induced remission by rapid depletion of plasma tryptophan. Arch Gen Psychiatry. 1990;47:411-418.
15. Smith KA, Fairburn CG, Cowen PJ. Relapse of depression after rapid depletion of tryptophan. Lancet. 1997;349:915-919.
16. Benkelfat C, Ellenbogen MA, Dean P, Palmour RM, Young SN. Mood-lowering effect of tryptophan depletion. Enhanced susceptibility in young men at genetic risk for major affective disorders. Arch Gen Psychiatry. 1994;51:687-697.
17. Harmer CJ, Bhagwagar Z, Cowen PJ, Goodwin GM. Acute administration of citalopram facilitates memory consolidation in healthy volunteers. Psychopharmacology (Berl). 2002;163:106-110.
18. King MV, Marsden CA, Fone KC. A role for the 5-HT(1A), 5-HT4 and 5-HT6 receptors in learning and memory. Trends Pharmacol Sci. 2008;29:482-492.
19. Beck AT. Cognitive Therapy – Nature and Relation to Behavior Therapy. Behav Ther. 1970;1:184-200.
20. Cuijpers P, van Straten A, Andersson G, van Oppen P. Psychotherapy for depression in adults: a meta-analysis of comparative outcome studies. J Consult Clin Psychol. 2008;76:909-922.
21. Harmer CJ, Goodwin GM, Cowen PJ. Why do antidepressants take so long to work? A cognitive neuropsychological model of antidepressant drug action. Br J Psychiatry. 2009;195:102-108.
22. Boureau YL, Dayan P. Opponency revisited: competition and cooperation between dopamine and serotonin. Neuropsychopharmacology. 2011;36:74-97.
23. Haslam N. The Discreteness of Emotion Concepts: Categorical Structure in the Affective Circumplex. Pers Soc Psychol Bull. 1995;21:1012-1019.
24. Snaith RP, Hamilton M, Morley S, Humayan A, Hargreaves D, Trigwell P. A scale for the assessment of hedonic tone the Snaith-Hamilton Pleasure Scale. Br J Psychiatry. 1995;167:99-103.
25. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960; 23:56-62.
26. Martinotti G, Sepede G, Gambi F, et al. Agomelatine versus venlafaxine XR in the treatment of anhedonia in major depressive disorder: a pilot study. J Clin Psychopharmacol. 2012;32:487-491.

Keywords: anhedonia; antidepressant; cognition; major depression; treatment