Falsification of memory in the absence of deceitfulness occurring in clear consciousness in association with amnesia

In Latin confabulari is 'to chat', fabula—'a tale', and fabulae!—'nonsense'. The confabulator indeed tells a story, but this story is not complete nonsense. In many cases it is more or less coherent and internally consistent, yet false in the "context named (Talland 1965; Moscovitch 1989). Typically, the account concerns the person who tells it, who is unaware of its memory deficit.

Confabulation in its broader meaning (definition 1) refers to a wide spectrum of phenomena. Some of these phenomena are within the normal range of the human behavioural repertoire. Minor, innocent confabulations may pop-up occasionally in some individuals in stressful social situations; they could be considered a sort of defence mechanism.1 Creative confabulations could contribute to mythology, literature, and the fine arts, as noted, for example, by Hobbes (1651):

Much memory, or memory of many things, is called experience. Againe, Imagination being only of those things which have been formerly perceived by Sense ... the imaging of the whole object, as it was presented to the senses is simple Imagination; as when one imagineth a man, or horse, which he hath seen before. The other is compounded; as when from the sight of a man at one time, and of a horse at another, we conceive in our mind a Centaure. So when a man compoundeth the image of his own person, with the image of the actions of an other man; as when a man imagins himselfe a Hercules, or an Alexander (which hap-peneth often to then that are much taken with reading of Romants) it is compound imagination, and properly but a Fiction of the mind.

Only pathological confabulations are of interest to us here. More specifically, only selected aspects of pathological confabulations. Hence antisocial personality disorder, which is expressed among others in deceitfulness (DSM-IV 1994), is not of our concern here; neither is Munchausen's syndrome, a psychiatric disorder in which an otherwise healthy individual seeks invasive medical treatment for feigned or self-induced symptoms (Ahser 1951).2 Left out is also confabulation in states of delirium and dense "dementia. To the student of memory, the most intriguing is confabulation in clear consciousness, which is adjunct to amnesia (definition 2). It fits the oxymoron 'honest lying'. Two types of confabulations are discerned in this case (Berlyne 1972; Kopelman 1987; Schnider etal. 1996a). One is 'momentary'or 'provoked'; the other is 'fantastic' or 'spontaneous'.3

Momentary or provoked confabulation is so called because it is fleeting and provoked by questions probing the "subject's memory. It is seen in amnesics, as well as in demented patients in the early stages of their

Fig. 18 A laboratory test that confabulators fail. Schnider and Ptak (1999) subjected spontaneous confabulators, nonconfabulating amnesics, and nonamnesic *controls, to a continuous Recognition task. In each run of the task, 80 pictures were presented one after the other on a computer screen for 2s each (A). Some of the items recurred (targets, marked by asterisk, total 28 in each run), whereas the other items did not recur (distractors, total 52 in each run). The subjects were requested to indicate item recurrence. Immediately after the completion of the first run, a second run was made, with a different selection of targets and distractors.At the beginning of the second run the subjects were asked to ignore the first run and indicate item recurrence only within the second run. Similarly, 5min after the second run, the third run was made, and 30min after the third run, the fourth run was made. As expected, spontaneous confabulators and nonconfabulating amnesics did not fare well on target detection (Ba); but most striking was the steep increase of false positive identification by the spontaneous confabulators (Bb). The experiment was replicated with nonsense instead of meaningful pictures (not shown), with similar results. These data were construed to indicate that the confabulators fail to suppress currently irrelevant memory traces. (C). The overlapping lesions in the brain of the spontaneous confabulators (a) and the noncofabulating amnesics (b). The basal forebrain and the medial orbitofrontal cortex were damaged in confabulators but not in nonconfabulating amnesics; in the latter, the lesions covered the posterior mediotemporal lobe and the dorsolateral prefrontal cortex. (Adapted from Schnider and Ptak 1999.)

Fig. 18 A laboratory test that confabulators fail. Schnider and Ptak (1999) subjected spontaneous confabulators, nonconfabulating amnesics, and nonamnesic *controls, to a continuous Recognition task. In each run of the task, 80 pictures were presented one after the other on a computer screen for 2s each (A). Some of the items recurred (targets, marked by asterisk, total 28 in each run), whereas the other items did not recur (distractors, total 52 in each run). The subjects were requested to indicate item recurrence. Immediately after the completion of the first run, a second run was made, with a different selection of targets and distractors.At the beginning of the second run the subjects were asked to ignore the first run and indicate item recurrence only within the second run. Similarly, 5min after the second run, the third run was made, and 30min after the third run, the fourth run was made. As expected, spontaneous confabulators and nonconfabulating amnesics did not fare well on target detection (Ba); but most striking was the steep increase of false positive identification by the spontaneous confabulators (Bb). The experiment was replicated with nonsense instead of meaningful pictures (not shown), with similar results. These data were construed to indicate that the confabulators fail to suppress currently irrelevant memory traces. (C). The overlapping lesions in the brain of the spontaneous confabulators (a) and the noncofabulating amnesics (b). The basal forebrain and the medial orbitofrontal cortex were damaged in confabulators but not in nonconfabulating amnesics; in the latter, the lesions covered the posterior mediotemporal lobe and the dorsolateral prefrontal cortex. (Adapted from Schnider and Ptak 1999.)

cognitive deterioration. For example, a subject asked to recognize a picture of the British Royal couple, might reply that they are celebrated movie stars, and go on fabricating details about the last movie in which they allegedly starred. This confabulation is momentary, does not significantly affect the subject's behaviour, and can be understood in terms of compensation for the deficient recollection; some nonamnesics and nonde-mented occasionally do just this, if weakness of character prevents them from simply admitting that they do not remember. In contrast, fantastic or spontaneous confabulation is unprompted, sustained, refers to wideranging narratives, may contain elements of grandiose, and has a persistent effect on the subject's behaviour. Spontaneous confabulation is not always fantastic. It could include sensible reminiscences that are out of their spatiotemporal context. Consider, for example, the patient who suffered from brain haemorrhage, and who for weeks afterwards confabulated about performing stocktaking in a certain store, although the last time he was supposed to do that was years earlier (Burgess and McNeil 1999). At the other side of the spectrum are confabulations whose source is more difficult to identify, as that of the Korsakoff's patient (*amnesia) recovering from a motorcycle accident in Britain, who came to believe she was in Royal Air Force and that her accident occurred while flying over France. Although over time she changed some details in the story, the flying theme remained prominent throughout for several months (Berlyne 1972).

The neuropathology of confabulation commonly involves damage to frontal structures (Figure 18), which could result from a variety of insults (Alexander and Freedman 1984; Stuss and Benson 1984; Shallice 1988; DeLuca and Cicerone 1991). These include rupture of aneurysm of the anterior communicating artery, causing damage the ventromedial frontal lobe;4 Korsakoff's syndrome (*amnesia); and physical injury. That frontal lobe amnesia is associated with confabulations suggests that some impairment of executive function is involved (*working memory). But this by itself still doesn't explain why confabulators confabulate.

So why do they? In searching for candidate neurological causes, it is first useful to *recall the distinction between spontaneous and provoked confabulation. As noted above, provoked confabulations may basically reflect a normal compensatory strategy (Kopelman 1987; Schnider et al. 1996a). Spontaneous confabulation, on the other hand, is much farther from normal, and therefore more interesting. Poor memory per se is not the reason, or at least is not a sufficient one, because many amnesics do not confabulate. Most investigators now believe that the problem is related to *retrieval.

Either the retrieval set is faulty to begin with, or the proper retrieval *cues become ineffective, or improper cues intrude, or the information gets mixed up in ecphory,5 or the *metamemory system does not monitor the retrieval process properly and fails to eliminate wrong results. The literature contains arguments for and against any of the above hypotheses.

A particularly fruitful hypothesis is that confabulation is a disorder of mnemonic chronology, and that the confabulated account is composed of bits of experiences taken out of their proper chronological and, hence also, factual context (Van der Horst, cited in Berlyne 1972; Schnider et al. 1996a,b). Even bizarre confabulations might assimilate information that has been encountered earlier, although not necessarily experienced, by the patient. Disorders of mnemonic chronology are probably insufficient to explain confabulation per se—many amnesics err in the chronology of experiences yet do not confabulate—but it might be a necessary condition for confabulation to occur. A variant of the mnemonic chronology deficit was proposed and tested by Schnider and Ptak (1999). They compared the *performance of confabulating and noncon-fabulating amnesics on a continuous *recognition task. In this task, the subjects had to identify recurrent pictures within ongoing test sessions (Figure 18). The con-fabulators made an increasingly growing number of false positive responses to pictures that were presented in the former but not in the ongoing session. It seemed that they recognized information encountered many minutes ago as if it had been presented in the present. This was construed to indicate that the confabulators lose the temporal label of stored information, and fail to suppress activated *internal representations even if these representations are irrelevant to current reality.

An intriguing spin-off of the aforementioned study is the suggestion that failure to recall properly may result from activation or reconstruction of too many memory traces, rather than from the lack of traces to retrieve. This mnemonic failure of certain amnesics reminds one of the severe problems that their antipodes, the hyper-amnesics, have in daily life (*mnemonics). In the storage (*metaphor) and recall of memory, as in so many other aspects of life, too much may be too much indeed.

Selected associations: Amnesia, Episodic memory, False memory, Recall, Retrieval

'On defence mechanisms in general, see *forgetting.

2A related syndrome is Munchausen's syndrome by proxy, which is illness fabricated by one person, usually a parent, in another, usually a child (Meadow 1977). These syndromes are called after Baron von Munchausen (1720-90), a retired German officer whose after dinner confabulations became a literary 'classic (Raspe 1785). It is noteworthy that Munchausen's syndromes might be facilitated by the spread of the Internet (Feldman 2000).

3The distinction between 'momentary' and 'fantastic' is not based on dissociation along the same 'dimension, whereas that between 'provoked' and 'spontaneous' is. Most authors now prefer the latter dichotomy. Both pairs of terms are, however, mentioned here because both are still used, sometimes interchangeably, in the literature. 4Aneurysm is a localized pathological dilatation of a blood vessel caused by congenital or acquired structural deficiency or inflammation of the vessel's wall.An aneurysm may rupture and cause haemorrhage, or it can become sufficiently large to displace and damage adjacent tissue.

5Ecphory is the actual act of retrieving or reconstructing the information, see 'retrieval.

Was this article helpful?

0 0

Post a comment