Associative recognition in a patient with selective hippocampal lesions and relatively normal item recognition.

Previous work (Mayes et al., Hippocampus 12:325-340, 2002) found that patient YR, who suffered a selective bilateral lesion to the hippocampus in 1986, showed relatively preserved verbal and visual item recognition memory in the face of clearly impaired verbal and visual recall. In this study, we found that YR's Yes/No as well as forced-choice recognition of both intra-item associations and associations between items of the same kind was as well preserved as her item recognition memory. In contrast, YR was clearly impaired, and more so than she was on the above kinds of recognition, at recognition of associations between different kinds of information. Thus, her recognition memory for associations between objects and their locations, words and their temporal positions, abstract visual items or words and their temporal order, animal pictures and names of professions, faces and voices, faces and spoken names, words and definitions, and pictures and sounds, was clearly impaired. Several of the different information associative recognition tests at which YR was impaired could be compared with related item or inter-item association recognition tests of similar difficulty that she performed relatively normally around the same time. It is suggested that YR's familiarity memory for items, intra-item associations, and associations between items of the same kind was mediated by her intact medial temporal lobe cortices and was preserved, whereas her hippocampally mediated recall/recollection of these kinds of information was impaired. It is also suggested that the components of associations between different kinds of information are represented in distinct neocortical regions and that initially they only converge for memory processing within the hippocampus. No familiarity memory may exist in normal subjects for such associations, and, if so, YR's often chance recognition occurred because of her severe recall/recollection deficit. Conflicting data and views are discussed, and the way in which recall as well as item and associative recognition need to be systematically explored in patients with apparently selective hippocampal lesions, in order to resolve existing conflicts, is outlined.

Remembering and knowing in a patient with preserved recognition and impaired recall.

ROB is a patient who has a severe deficit in recalling recently presented verbal material following rupture and repair of an anterior communicating artery aneurysm [Hanley JR, Davies ADM, Downes J, Mayes A. Cognitive Neuropsychology 1994;11:543-78; Hanley JR, Davies ADM. In: Parkin A, editor. Case Studies in the Neuropsychology of Memory. Hillsdale, NJ: Lawrence Erlbaum, 1997. p. 111-26]. Despite this, her performance on tests of recognition memory is comfortably within the normal range. In the present series of experiments, we investigated whether or not ROB's performance on tests of recognition memory might be associated with a disproportionately large number of correct decisions made on the basis of familiarity rather than contextual retrieval [e.g. Mandler G. Psychological Review 1980;87:252-71]. Contrary to this hypothesis, the results showed that ROB made a high proportion of remember decisions relative to know decisions in recognition [cf. Gardiner JM. Memory & Cognition 1988;16:309-13] and produced a high recollection score when conscious recollection and familiarity were placed in opposition to one another [cf. Jacoby LL, Woloshyn V, Kelley C. Journal of Experimental Psychology: General 1989;118:115-25.]. ROB's recognition memory performance therefore appears to be qualitatively as well as quantitatively similar to that found in the normal population. As ROB has suffered damage to both the fornix and the anterior thalamus, the results of the present study are consistent with the claim that damage to the extended hippocampal system has a much more severe effect on recall than on recognition [Aggleton JP, Shaw C. Neuropsychologia 1996;34:51-62; Aggleton JP, Saunders RC. Memory 1997;5:49-71]. The present results provide no support, however, for the additional suggestion [Aggleton JP, Brown MW. Behavioral and Brain Sciences 1999;22:425-56.] that the extended hippocampal system is necessary for recognition memory decisions that are based on contextual retrieval.

An fMRI study of stimulus equivalence.

In order to study brain activation during the formation of equivalence relations, 12 subjects underwent fMRI during matching-to-sample (MTS) tests of (1) previously trained arbitrary relationships between iconic stimuli and the untrained, emergent relations of (2) symmetry, (3) transitivity, and (4) symmetry with transitivity, plus a test of verbal fluency (VF). Brain activation was similar in all MTS tasks and in the VF task. In particular, both types of task activated dorsolateral prefrontal cortex (DLPFC) and posterior parietal cortex bilaterally. However VF, but not the MTS tasks, activated Broca's area. In three of the four MTS tasks, behavioural accuracy was significantly correlated with left lateralisation of DLPFC activity. Brain activation patterns during equivalence thus resembled those involved in semantic processing underlying language, without involving regions concerned with the simple sub-vocal articulation of stimulus names.

Memory for single items, word pairs, and temporal order of different kinds in a patient with selective hippocampal lesions.

One kind of between-list and two kinds of within-list temporal order memory were examined in a patient with selective bilateral hippocampal lesions. This damage disrupted memory for all three kinds of temporal order memory, but left item and word pair recognition relatively intact. These findings are inconsistent with claims that (1) hippocampal lesions, like those of the medial temporal lobe (MTL) cortex, disrupt item and word pair recognition, and that (2) hippocampal lesions disrupt temporal order memory and item recognition to the same degree. Not only was word pair recognition intact in the patient, but further evidence indicates that her recognition of other associations between items of the same kind is also spared so retrieval of such associations cannot be sufficient to support within-list temporal order recognition. Rather, as other evidence indicates that the patient is impaired at recognition of associations between different kinds of information, within-list (and possibly between-list) temporal order memory may be impaired by hippocampal lesions because it critically depends on retrieving associations between different kinds of information.

Long live Proust: the odour-cued autobiographical memory bump.

The autobiographical memory bump is an increase in the frequency of reported autobiographical memories (AMs) from a particular age range, and has been reported by numerous investigators (for reviews, see Conway, M. A. & Rubin, D. C. (1993). The structure of autobiographical memory. In A. F. Collins, S. E. Gathercole, M. A. Conway & P. E. Morris, Theories of memory. Hove: Lawrence Erlbaum; Rubin, D. C., Rahhal, T. A. & Poon, L. W. (1998). Things learned in early adulthood are remembered best. Memory and Cognition, 26, 3-19) to occur in the second and third decades of life. Invariably, word labels have been used as AM cues but, given that a popular interpretation of the anecdotal 'Proust phenomenon' is that odours evoke AMs which are especially aged, we wondered if a different pattern in the AM bump might emerge if AMs were cued by odours rather than labels. Here we report an attempt to substantiate this aspect of the 'Proust phenomenon' by comparing the distributions of memories across the lifespan when cued by odour and label. Data showed that, in line with previous studies, the bump for label cues was found to peak between ages 11 and 25 years and was confirmed to be quadratic in form. In contrast, the odour-cued memory distribution peaked at 6-10 years and decreased linearly thereafter. In the earliest age interval, 6-10 years, the proportion of AMs retrieved in response to odour cues was significantly greater than that for the label cues. These results provide empirical support for the Proust phenomenon, and have more general implications for the structure and age distribution of stored AMs.

Odour-evoked autobiographical memories: psychological investigations of proustian phenomena.

Folk wisdom dictates that odours are especially powerful reminders of autobiographical experience, an effect which has become known as the Proust phenomenon. This paper reviews the relevant literature to determine whether there is any substantive evidence to support this view. Different methodologies have been adopted in addressing this issue, but the most revealing and ecologically valid have been the few studies which have examined naturally formed autobiographical memories. From these data, there is at least preliminary evidence that olfactory stimuli can cue autobiographical memories more effectively than cues from other sensory modalities. Explanations for these effects can be invoked from accepted principles in contemporary cognitive psychology.

Intellectual, mnemonic, and frontal functions in dementia with Lewy bodies: A comparison with early and advanced Parkinson's disease.

Both Parkinson's disease (PD) and dementia with Lewy bodies (DLB) share a common neuropathological marker, the presence of Lewy bodies in brain stem and basal forebrain nuclei. DLB, in addition, is associated with Lewy bodies in the neocortex, and, in it's more common form, with Alzheimer-type pathological markers, particularly amyloid plaques. Published neuropsychological studies have focused on the differential profiles of DLB and Alzheimer's disease (AD). However, it is presently unclear whether DLB should be classified as a variant of AD or PD. In the present study we compare a healthy age-matched control group with three groups of patients, one with DLB, and two with PD. One of the PD groups was early in the course (PD-E) and the second, more advanced group (PD-A), was matched on severity of cognitive impairment with the DLB group. The results show that DLB was associated with a different pattern of neuropsychological impairment than the PD-A group, particularly in tests believed to be mediated by prefrontal cortical regions.

Do amnesics forget colours pathologically fast?

We tested amnesic and control subjects on a task which required the recognition of single, difficult to name colours, after delays ranging from 7 seconds to 120 seconds after performance of the two subject groups had been matched at the shortest delay by giving the amnesic patients longer study time. The amnesic patients showed abnormally fast forgetting over the two minute period. Furthermore, a subgroup of nine subjects with presumed damage to midline diencephalic structures (Korsakoff's syndrome) were found to forget as fast as a group of six subjects with presumed medial temporal lobe damage (herpes simplex encephalitis). These results contrast both with studies using the Huppert and Piercy procedure and those using the Brown-Peterson task, none of which have shown convincing evidence of accelerated forgetting in medial temporal lobe or diencephalic amnesia.

Does context discriminate recollection from familiarity in recognition memory?

Five experiments were conducted to examine subjects' ability to make contextual judgements about recognized items for which they report recollective experience or only familiarity within the context of the experiment. In the first four experiments, subjects were able to make judgements of the spatiotemporal context of items that were accompanied by recollective experience significantly better than for items they merely found familiar. In only one of the four studies did subjects display above-chance performance on spatiotemporal judgements for merely familiar items. A fifth experiment examined the frequency with which subjects report the presence of different kinds of contextual knowledge during a standard recognition experiment. All aspects of contextual knowledge were reported with higher frequencies for recollected items than for items only found familiar, although no single contextual feature was strongly associated with recollective experience. Thus, the five studies together provide converging evidence for the validity of the "recollect-know" distinction in recognition memory and supplement studies that have already demonstrated that the two kinds of response are dissociable. The implications of these data for group comparisons of memory-impaired patients, and the role of context in recognition memory are discussed.

Stem-completion priming in Alzheimer's disease: the importance of target word articulation.

Stem-completion priming performance in patients with Alzheimer's type dementia (DAT) was explored in three experiments in which both the standard repetition priming effect and a novel indirect form of priming, cohort priming, were measured. In the first experiment, in which study stimuli were words, both priming effects were found to be markedly attenuated in the DAT group. In the second experiment, the study stimuli were specially constructed nonwords, and it was found that cohort priming was present at normal levels in the DAT group. In a third experiment we tested the specific hypothesis that the requirement to overtly articulate target stimuli during the study phase was critical for the appearance of normal cohort priming in the DAT group in Experiment 2, and also for the normal levels of repetition priming which have been reported in some published studies. Two encoding conditions were compared, one in which subjects simply had to read aloud the target words and a second in which subjects were required to make evaluative (pleasantness) ratings for each of the target words (identical to that used in Experiment 1). Stem-completion priming performance following the latter condition was significantly attenuated in the DAT group relative to a healthy control group, but following the "read aloud" encoding condition, normal levels of repetition and cohort priming were observed. It is suggested that the most fruitful approach to understanding the performance of DAT subjects on lexical repetition priming tasks will involve a detailed analysis of language functions and how they interact with other, possibly mnemonic, processes in the generation of primed responses.

Postoperative apnea in former preterm infants after inguinal herniorrhaphy. A combined analysis.

BACKGROUND: Controversy exists as to the risk for postoperative apnea in former preterm infants. The conclusions of published studies are limited by the small number of patients. METHODS: The original data from eight prospective studies were subject to a combined analysis. Only patients having inguinal herniorrhaphy under general anesthesia were included; patients receiving caffeine, regional anesthesia, or undergoing other surgical procedures were excluded. A uniform definition for apnea was used for all patients. Eleven risk factors were examined: gestational age, postconceptual age, birth weight, history of respiratory distress syndrome, bronchopulmonary dysplasia, neonatal apnea, necrotizing enterocolitis, ongoing apnea, anemia, and use of opioids or nondepolarizing muscle relaxants. RESULTS: Two hundred fifty-five of 384 patients from eight studies at four institutions fulfilled study criteria. There was significant variation in apnea rates and the location of apnea (recovery room and postrecovery room) between institutions (P < 0.001). There was considerable variation in the duration and type of monitoring, definitions of apnea, and availability of historical information. The incidence of detected apnea was greater when continuous recording devices were used compared to standard impedance pneumography with alarms or nursing observations. Despite these limitations, it was determined that: (1) apnea was strongly and inversely related to both gestational age (P = 0.0005) and postconceptual age (P < 0.0001); (2) an associated risk factor was continuing apnea at home; (3) small-for-gestational-age infants seemed to be somewhat protected from apnea compared to appropriate- and large-for-gestational-age infants; (4) anemia was a significant risk factor, particularly for patients > 43 weeks' postconceptual age; (5) a relationship to apnea with history of necrotizing enterocolitis, neonatal apnea, respiratory distress syndrome, bronchopulmonary dysplasia, or operative use of opioids and/or muscle relaxants could not be demonstrated. CONCLUSIONS: The analysis suggests that, if it is assumed that the statistical models used are equally valid over the full range of ages considered and that the average rate of apnea reported across the studies analyzed is accurate and representative of actual rates in all institutions, the probability of apnea in nonanemic infants free of recovery-room apnea is not less than 5%, with 95% statistical confidence until postconceptual age was 48 weeks with gestational age 35 weeks. This risk is not less than 1%, with 95% statistical confidence, for that same subset of infants, until postconceptual age was 56 weeks with gestational age 32 weeks or postconceptual age was 54 weeks and gestational age 35 weeks. Older infants with apnea in the recovery room or anemia also should be admitted and monitored. The data do not allow prediction with confidence up to what age this precaution should continue to be taken for infants with anemia. The data were insufficient to allow recommendations regarding how long infants should be observed in recovery. There is additional uncertainty in the results due to the dramatically different rates of detected apnea in different institutions, which appear to be related to the use of different monitoring devices. Given the limitations of this combined analysis, each physician and institution must decide what is an acceptable risk for postoperative apnea.

Do amnesics forget faces pathologically fast?

In 1978, Huppert and Piercy introduced a general method for comparing forgetting rates across groups differing in their baseline memory performance. The method has since become a standard for measuring rate of forgetting in amnesia. Using this method, amnesic subjects with presumed damage to midline diencephalic structures have consistently been reported to forget at a normal rate whereas patients with medial temporal lobe damage have sometimes been reported to forget pathologically fast. Conclusions about amnesic forgetting rates using Huppert and Piercy's procedure, however, are unsafe because the matching procedure results in the shortest mean item-presentation-to-test delay being longer in amnesics than control subjects. A further problem with previous work is that frequently the shortest delay at which performance is measured is 10 minutes. An alternative procedure to Huppert and Piercy's is outlined which eliminates the matching confound. An experiment was carried out using this procedure with face stimuli, and with amnesic and control performance matched immediately following study, and then tested at delays of 5, 12, and 30 minutes. Pathologically fast forgetting was observed in a group of 19 amnesics over the first 5 minutes, but between 12 and 30 minutes their controls forgot faster so that the two groups had forgotten the same amount after 30 minutes. A subgroup of nine Korsakoff patients, with probable damage to midline diencephalic structures, showed a similar abnormal forgetting pattern to the remaining 10 amnesics, some of whom had medial temporal lobe damage. A retroactive interference condition was also included for the 12 minute condition at which delay patient and control recognition was mildly and equivalently disrupted. For unknown reasons perhaps related to a storage abnormality, amnesics lose face recognition memory sooner in the first 30 minutes of forgetting than do normal people, who show accelerated forgetting later so as to match patients after 30 minutes delay.

Two tests for assessing remote public knowledge: a tool for assessing retrograde amnesia.

Guidelines for the construction and development of tests of remote memory are provided. These guidelines follow from methodological considerations and from certain theoretical issues we believe that remote memory tests should address if they are to further our understanding of amnesia. These include: the pattern of temporal gradient; memory for different broad classes of information such as famous names and famous events; and recall and recognition of different types of detail such as contextual and non-contextual information associated with remote memory. With respect to the first of these, we argue that it is impossible to fully evaluate competing hypotheses about temporally graded retrograde amnesia unless items are selected from at least two difficulty levels, and from at least two time periods. The construction of two equivalent tests of remote memory, one for famous people, the other for famous events, is described, and the theoretical issues listed earlier were addressed using these tests in a study comparing the remote memory of 13 organic amnesics and 13 age- and IQ-matched controls. The principal findings were as follows. There was only weak evidence that amnesics' remote memory for famous people and events was temporally graded over a 20 year period, and recognition memory and memory for non-contextual detail were equally impaired. Mayes, Meudell, and Pickering's (1985) context memory deficit hypothesis was supported, however, because amnesics were more impaired in their ability to date recognised items. In addition, although there was little support for a temporally graded retrograde impairment on the recognition question, there was much clearer evidence for the temporal dating question. Also, one patient, SR, showed a selective deficit in dating recognised names and events from the 1980s. Finally, the correlations between measures of anterograde and retrograde amnesia were all non-significant. Suitable items for the construction of remote memory tests probing the four decades 1950s to 1980s are presented. A subset of these, from the 1970s and 1980s, were shown to have satisfactory psychometric qualities, and can therefore be recommended for more widespread clinical use, although more extensive normative data than that provided here will be necessary.