Old dogs with new tricks: Detecting accelerated long-term forgetting by extending traditional measures.

YEBEH-04201; No of Pages 7 Epilepsy & Behavior xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh

Old dogs with new tricks: Detecting accelerated long-term forgetting by extending traditional measures Laurie A. Miller a,b,⁎, Emma Flanagan c, Annu Mothakunnel d, Armin Mohamed a, Zoe Thayer a,b a

Institute of Clinical Neurosciences, Royal Prince Alfred Hospital, Camperdown, NSW, Australia ARC-Centre of Excellence in Cognition and Its Disorders, University of Sydney, NSW, Australia Neuroscience Research Australia, Sydney, NSW, Australia d ARC-Centre of Excellence in Cognition and Its Disorders, Macquarie University, NSW, Australia b c

a r t i c l e

i n f o

Article history: Received 3 December 2014 Revised 19 January 2015 Accepted 21 January 2015 Available online xxxx Keywords: Epilepsy Memory Normative data Recall

a b s t r a c t Accelerated long-term forgetting (ALF) is a condition in which normal memory performance is displayed after short delays, but significant memory loss is detected when memory is tested after several days or weeks. This condition has been reported in patients with epilepsy, but there are few normative scores available for its detection in clinical practice. In the present study, we assessed 60 healthy control subjects 18–60 years of age on three memory measures [Rey Auditory Verbal Learning (RAVLT), Logical Memory (LM), and Aggie Figures] at delays of 30 min and 7 days. With these normative values, we determined cutoff scores to look for ALF and then categorized the performance of 15 patients with focal epilepsy on the same tasks. Seven of the patients showed ALF, and, in four of these, no other memory deficits (i.e., deficits at 30 min on at least one task) were detected. Of the several demographic and epilepsy factors examined, only higher estimated IQ and older age predicted ALF (and only on one task: RAVLT). The findings provide a useful set of data to be applied in the clinic and some insight into the factors that influence retention within the first week. © 2015 Elsevier Inc. All rights reserved.

1. Introduction Accelerated long-term forgetting (ALF) [1,2], also known as longterm amnesia [3–5], is a condition in which a person shows normal memory in the short run (i.e., 20–30 min postpresentation) but significant loss of material after longer delays (e.g., days or weeks). Because neuropsychological tests usually only involve assessment of memory at shorter intervals, this condition cannot be detected under standard conditions. However, when special efforts are made to measure memory over longer periods, ALF has frequently been reported in studies of patients with both temporal lobe epilepsy (TLE) [6–8] and transient epileptic amnesia (TEA) [9]. Recently, it has also been identified in patients with extratemporal foci [10]. Furthermore, patients' complaints with regard to everyday memory have been found to better correlate with impairments in long-term delayed recall than with deficits in recall at 30 min [9,11,12], emphasizing the importance of capturing longer term memory deficits as part of the clinical assessment. Abbreviations: ALF, accelerated long-term forgetting; EEG, electroencephalography; ETE, extratemporal epilepsy; FSIQ, Full Scale Intelligence Quotient; NC, normal control; MRI, magnetic resonance imaging; RAVLT, Rey Auditory Verbal Learning Test; TEA, transient epileptic amnesia; TOPF, Test of Premorbid Functioning; TLE, temporal lobe epilepsy; WMS, Wechsler Memory Scale. ⁎ Corresponding author at: Neuropsychology Unit, Royal Prince Alfred Hospital, Camperdown, NSW 2050, Australia. Tel.: +61 2 9515 7816; fax: +61 2 9515 7564. E-mail address: [email protected] (L.A. Miller).

Many of the previous investigations of ALF have employed “in-house” material specifically designed to address particular study questions (e.g., [5,12–18]). In the subset of studies that have tested long-term performance on routinely presented clinical material [e.g., Rey Auditory Verbal Learning Test (RAVLT), Rey Complex Figure Test, and Wechsler Memory Scale subtests], normative values from matched control subjects either are based on a small sample (e.g., [12,19–21]) or have yielded no significant differences between the control group and the patient group [22]. However, in 1997, Geffen et al. [23] provided norms for the RAVLT word list, including mean recall scores after 7 days, from a large normative sample. They found that younger (20–39 years of age) and older (40–59 years of age) groups performed similarly on both learning and 7-day delayed recall. They also determined that long-term recall results obtained over the telephone were not different from those obtained in face-to-face testing. In a recently published critical review of ALF studies, Elliott et al. [24] concluded that the creation of a set of standardized clinical ALF tests that involve both verbal stimuli and nonverbal stimuli and have suitable sensitivity would be ideal. Although the RAVLT and the Logical Memory (prose passages) subtests of the Wechsler Memory Scale are commonly used to assess verbal memory in epilepsy centers (at least at short delays), there is more variability in the tests used to assess nonverbal memory [25,26]. Jones-Gotman and colleagues [26–28] introduced the use of lists of abstract designs (e.g., Aggie Figures) for learning and retention as a visual analogue to the RAVLT. For this sort of material,

http://dx.doi.org/10.1016/j.yebeh.2015.01.024 1525-5050/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: Miller LA, et al, Old dogs with new tricks: Detecting accelerated long-term forgetting by extending traditional measures, Epilepsy Behav (2015), http://dx.doi.org/10.1016/j.yebeh.2015.01.024

2

L.A. Miller et al. / Epilepsy & Behavior xxx (2015) xxx–xxx

patients with large right hippocampal removals show deficits in (1) learning over the first few presentation trials and (2) percent retention over a 24-hour period [28]. Longer delays were not considered. The first goal of this study was to provide normative data for long-term (i.e., 7 days) recall of three memory tests: RAVLT, WMS-IV Logical Memory, and Aggie Figures. Interestingly, in healthy subjects, sex effects have been found on tests of anterograde verbal memory. On the RAVLT, Geffen et al. [23] found that women outperformed men in memory for words at both 30-min and 7-day recall. Women also tend to show better memory for stories (at least until 30-min delay) [21,29], but, to our knowledge, sex differences in story retention over longer intervals have not been studied. With regard to memory for visual material, in one study, women performed better than men at remembering the position of an object in an array [30], but no one has found evidence of sex differences on tasks involving memory for abstract designs. For the present study, we considered possible effects of sex, along with age and education level when determining normative values for the target tasks. With respect to identifying ALF in patients with focal epilepsy, previous investigations have tended to compare the mean score achieved by the patient group with that achieved by the control group to determine whether the patients with epilepsy (as a group) show pattern characteristic of ALF (e.g., [7,19,31,32], see also [19]). Hence, it is not clear whether (and how often) one would detect ALF in individual patients with focal epilepsy using cutoff scores. This could be done by establishing a quantitative definition of ALF (i.e., normal memory initially with greater-than-expected loss over time) based on normative values. The second goal of this study was to examine rates of ALF in patients with focal epilepsy. Several factors have been found to contribute to ALF. In a review paper [33], we concluded that older age and higher IQ seem to be linked with this diagnosis. In adults with epilepsy, ALF has primarily been identified in those with a temporal lobe focus [3–5,8,34–36] but is not necessarily linked to hippocampal abnormalities [9,22,37]. Consistent with this, Lah et al. [32] reported a steep decline in memory over a 24-hour period in patients with TLE with hippocampal pathology on MR imaging but a more gradual decline in memory over several days in patients with TLE with no hippocampal abnormalities. Side of lesion has also not been found to predict its occurrence in a material-specific fashion [22,31]. That is, there is little evidence that left-sided epilepsy results in selective ALF for verbal material and that right-sided epilepsy causes ALF for visual material (which would be suggested from patterns seen after shorter intervals) [38–41]. Instead, there is some indication that patients with left-sided epilepsy have more pervasive memory deficits (i.e., affecting both verbal retention and nonverbal retention), whereas patients with right-sided foci have shown ALF limited to memory for designs (sparing long-term verbal memory) [6,12]. The third aim of the present study was to consider the possible influence of these factors in a prospective sample of patients with focal epilepsy who presented to our clinic for assessment. Finally, we looked at the overlap and the differences between patients found to have ALF and those who showed memory defects at 30 min. We hypothesized that the presence of a hippocampal lesion might be more likely in the latter case.

groups. In addition to age, subjects were divided according to gender (male or female) and education level (less than or equal to 12 years of education or post-high school education). Participants with any post-high school education were not divided further because analyses indicated no differences between those with postgraduate qualifications and those without postgraduate qualifications on any of the memory measures. Two NC participants failed to return the Aggie Figures response sheet after one week, so those data points are missing. A prospective sample of 23 patients thought to have epilepsy who presented to the seizure clinic for neuropsychological evaluation also underwent assessment on all three long-term memory measures described below. These subjects met the same criteria listed above with the exception of having a history of seizures. Nine of these patients were eliminated from this study either because epilepsy was not confirmed or because they had primary generalized as well as focal-onset seizures. The 15 patients with EEG-confirmed focal epilepsy who were tested in this study included 8 women and 7 men between 18 and 61 years of age (mean = 37). See Table 1 for information on the distribution of the NC and patient participants with regard to demographics. On the basis of imaging, EEG studies, and clinical history, the side and site of epileptic focus were determined without knowledge of the neuropsychological results. There were 9 patients with TLE and 6 patients with extratemporal epilepsy (ETE). Eight patients had left-sided foci, 6 patients had right-sided foci, and, in 1 patient, lateralization of seizure focus had not been determined. Five patients had hippocampal lesions, 1 patient had undergone hippocampal removal as part of a temporal lobectomy, and the other 4 patients had hippocampal sclerosis. The remaining 10 patients had no visible hippocampal abnormality on MRI. A Supplemental Table is provided with all of the patient details. All subjects gave their informed consent, and the project was approved by both the University of Sydney and the Royal Prince Alfred Hospital Human Ethics Committee.

2. Method

2.2.1. Rey Auditory Verbal Learning Test (RAVLT) The Rey Auditory Verbal Learning Test is a well-known clinical measure of verbal learning and memory [42]. A 15-item word list (List A) is presented over five learning trials. In each trial, the list is read aloud to the subjects, and they are asked to recall as many words as possible. A distractor list of 15 items (List B) is then read by the examiner and recalled by the subject in a similar fashion (but only once), and an unprompted (“immediate”) recall of List A follows. Twenty to 30 min later, delayed recall of List A is requested. For this study, subjects were also called 7 days later, and long-term recall of List A was collected over the telephone. Scores used in this evaluation were the recall at 30 min and the Percent Change score: [(30-min recall − 7-day recall) / 30-min recall] × 100.

2.1. Subjects Sixty healthy normal control (NC) subjects were recruited as a sample of convenience from amongst friends and relatives of the three investigators as well as relatives of patients seen at the Neuropsychology Unit at Royal Prince Alfred Hospital. Subjects were asked to participate if they met the following criteria: spoke English as their best language and had no psychiatric or neurological history (including no previous seizures). We intentionally sought equal numbers of males and females evenly distributed in older (40–60 years of age) and younger (18–39 years of age)

2.2. Materials Three memory tests were used in this investigation. The stimuli for all three were administered as part of a larger study in a session that lasted approximately 90 min. In all three cases, the intervals between immediate recall and 30-min delayed recall were filled with other tests, as typically occurs in a clinical assessment. The intervening material was of a different modality than the test material. Tests were administered in the order presented below. Following standard protocol, subjects were not forewarned about any of the recalls. Although often included in a clinical assessment, for this study, recognition memory was not tested at 30 min to avoid the representation of the material serving as a reminder. At the end of the standard in-office assessment, arrangements were made for a convenient time and telephone number at which a followup interview could be performed 7 days post-assessment, and the subject was provided with a self-addressed stamped envelope and a blank piece of paper (unbeknownst to them, for the long-term Aggie Figures recall).



3

Table 1 Distribution of the NC and patient participants according to age, sex, and level of education. Ages 18–29

NC participants (n) n with b12 years of education n with N12 years of education Patient participants (n) n with b12 years of education n with N12 years of education

Ages 30–39

Ages 40–49

Ages 50–61

Male

Female

Male

Female

Male

Female

Male

Female

8 5 3 1 1 0

10 0 10 2 0 2

7 1 6 2 2 0

4 2 2 4 3 1

5 2 3 3 1 2

10 1 9 1 0 1

8 0 8 1 0 1

8 0 8 1 0 1

2.2.2. WMS-IV Logical Memory This verbal memory measure from the Wechsler Memory Scale — IV [43] involves two stories that are read aloud one at a time and recalled by the subject immediately and then again after 25–35 min. Seven days later, recall was tested over the telephone. Consistent with standard protocol, if the subject could not recall any details from a story at the delayed (or long-term) recall, a prompt was provided. Recalls from the two stories were added together. Scores used in this evaluation were the recall at 30 min and the Percent Change score: (30-min recall − 7-day recall) / 30-min recall.

2.2.3. Aggie Figures This test of memory for a set of 15 abstract line drawings was created and provided by Dr. Jones-Gotman and colleagues [27]. It is a visual analogue of the RAVLT and is presented in a similar fashion. Each line drawing of List A is shown to the subject for a few seconds, one after another, using a ring-bound booklet. After seeing all 15 items, subjects are provided with a blank sheet of paper and are asked to draw as many as possible. This is repeated for 5 learning trials, and then a distractor list (List B) of 15 new drawings is presented and recalled. Immediate recall and 20- to 30min delayed recall of List A follow. At the time of the phone call seven days later, subjects were instructed to draw again all the figures they could remember and then post back the sheet in the supplied envelope. Scores used in this evaluation were the recall at 30 min and the Percent Change score: [(30-min recall − 7-day recall) / 30-min recall] × 100.

3. Results 3.1. Normative scores For the NC group, we found no significant main effects of age for either score on any of the three target measures. That is, older (40– 60 years of age) and younger (18–39 years of age) subjects performed similarly on the 30-min delayed recall and in Percent Change scores on all three types of material. Furthermore, there were no significant Age × Education or Age × Sex interactions. In addition, Pearson correlation tests indicated no significant relationship between age and test scores (delayed recall or Percent Change) for any of the three measures. Hence, subjects of all ages were combined for the following analyses. The main effect of Sex was evident on the 30-min delayed recall scores from both of the verbal memory measures (RAVLT 30 min: F(1,56) = 5.0, p = .03; LM 30 min: F(1,56) = 4.4, p = .04). In both cases, women recalled more words compared with men (see Tables 2 and 3). In contrast, men and women performed similarly in the 30-min recall of the Aggie Figures. The two sexes showed similar Percent Change scores on all three measures (i.e., no significant main effect of Sex). The mean recall score at 30 min was not significantly different for the two levels of education on any of the three measures. With respect to the Percent Change score, the main effect of Education was only apparent on Aggie Figures (F(1,54) = 9.9, p = .003); those with a higher level of education had lower Percent Change over time (see Table 4).

3.2. Detection of ALF in patients 2.2.4. Test of Premorbid Functioning (TOPF) The TOPF [44] was used to provide an estimate of each participant's Full Scale Intelligence Quotient (FSIQ). The TOPF is based on reading pronunciation ability for words that have irregular grapheme-tophoneme translation. Participants are presented with a card that lists 70 words. They are asked to read aloud from the word card until they have completed the list or mispronounced 5 words consecutively. Raw scores are then converted to provide Wechsler Adult Intelligence Scale — IV FSIQ estimates.

2.3. Statistical analyses and case evaluation First, we used two-way ANOVAs involving either Age (younger; older) × Sex or Age × Education Level to look for differences in the NC group on the two scores for each of the three measures. Second, we determined which patients showed ALF by having normal recall at 30 min (z N −1.64) and a greater-than-normal loss of memory by 7 days as indicated by the Percent Change score (z N 1.64). Third, we explored the influence of factors previously linked with ALF using Fisher exact tests to consider categorical variables (e.g., the presence of a hippocampal lesion and side of epileptic focus) and regression analyses for continuous demographic and clinical variables. Fourth, we compared the rates and characteristics of patients with ALF with the rates and characteristics of patients with memory deficits at 30 min. In all cases, results reaching significance at p b .05 are reported.

As a group, the 15 patients did not differ from the control group in sex distribution or mean age, but they averaged fewer years of education (NC mean = 15.1; patient mean = 13.1) (t(73) = 3.3, p = .001) and lower estimated FSIQ (NC mean = 105; patient mean = 96) (t(73) = 2.7, p = .01). Overall, the patients averaged significantly lower scores than the NC group on all three mean 30-min recall scores (RAVLT: t(73) = 3.7, p b .001; LM: t(73) = 2.5, p = .02; Aggie Figures: t(73) = 3.0, p = .004). Furthermore, as a group, patients averaged higher Percent Change scores than the NC group on both verbal measures (RAVLT: t(73) = 3.9, p b .001; LM: t(73) = 2.9, p = .004) but not on Aggie Figures. The mean scores for the two groups are not discussed in further detail because the purpose of this study was to detect ALF in individuals; in other words, to identify those cases who

Table 2 Norms for RAVLT. 30-min delay (max = 15) Sex N Mean score (SD) Cutoff score (z = −/+1.64) For ALF

Male 28 10.5 (3.0) 5.6

Female 32 12.3 (2.4) 8.4

Performs above this level

Percent Change score (max = 100) [(30 min − 7 days) / 30 min] × 100 Male + female 60 36.3 (21.8) 72.1 Shows greater change score


4


Table 3 Norms for Logical Memory. 30-min delay (max = 50) Sex Male Female N 28 32 Mean score (SD) 22.0 (8.2) 26.9 (7.3) Cutoff score 8.6 14.9 (z = −/+1.64) For ALF Performs above this level

Percent Change score (max = 100) [(30 min − 7 days) / 30 min] × 100 Male + female 60 29.2 (20.6) 63.0 Shows greater change score

performed normally initially but showed a significant drop after a oneweek interval. An individual's performance was classified as showing evidence of ALF when the 30-min delayed recall score was within normal limits (z N − 1.64) and when the Percent Change Score was impaired (i.e., z N 1.64). Table 5 shows the details of the patients who met this criterion on each of the three memory measures. We can observe that all three tasks proved sensitive to ALF, with slightly more patients showing the ALF pattern on the RAVLT than on the other two measures. Approximately half of our patient sample (i.e., 7/15) demonstrated ALF, but, somewhat surprisingly, only one patient (JM) showed ALF on more than one test. Chi-square analysis indicated a significant difference between the percentage (47%) of patients and the percentage (17%) of NC subjects who demonstrated a pattern of ALF on at least one test (X2 = 6.2, p = .03). 3.3. Factors contributing to ALF Consideration of the patient characteristics shown in Table 4 indicates that when ALF is defined as normal recall at 30 min and greater than normal decline by 7 days, the pattern was predominantly seen in, though not completely limited to (i.e., case IP), patients with TLE. However, a Fisher exact test comparing this distribution with the TLE/ETE distribution in the overall sample did not reach significance for this relatively small sample. As can be seen in Table 4, an epileptic focus in either the left or the right hemisphere could result in ALF. Although there were slightly more patients with left-sided foci than patients with right-sided foci (i.e., 5 versus 2), again, this distribution did not differ significantly from the distribution by side of epilepsy focus in the overall group of 15 patients. Furthermore, there was no evidence of material-specific ALF associations with side of epileptic focus; there were patients with both right-sided and left-sided foci who showed ALF on the verbal tasks. And, on the visual task (Aggie Figures), the two patients who showed ALF had left-sided foci. Similarly, there was no indication that a hippocampal lesion was necessary for ALF to be found. The percentage (57%) of patients with a hippocampal lesion in Table 4 did not differ from their representation in the total sample (i.e., 33% with a hippocampal lesion). The sex distribution in Table 4 (i.e., patients who had ALF) and the sex distribution for the overall patient sample were also not significantly different. Univariate regression analyses involving the 15 patients indicated that older age (r2 = .56, β = .75, t = 4.0, p = .001) and higher estimated IQ scores (r2 = .33, β = .57, t = 2.5, p = .03) predicted a Table 4 Norms for Aggie Figures.

Education N Mean score (SD) Cutoff score (z = −/+1.64) For ALF

30-min delay (max = 15)

Percent Change score (max = 100) [(30 min − 7 days) / 30 min] × 100

All 60 11.3 (3.1) 6.2

b12 years 11 44.9 (29.0) 92.5

N12 years 47 27.7 (19.2) 59.2

Performs above this level Shows greater change score

diagnosis of ALF on the RAVLT (but not on the other two tests). The number of years of education was not predictive of a diagnosis of ALF on any of the three tests. In stepwise regression analyses, both age and IQ remained significant for ALF on the RAVLT (r 2 = .75; Age: β = .66, t = 4.0, p = .001; TOPF: β = .44, t = 3.0, p = .01). Duration of seizures was not a significant predictor of ALF on any of the three tests. No subject required standard prompting to remember the Logical Memory prose passages at 30-min delay, but 3 patients required prompting at the 7-day delay. In spite of prompting, one of these cases (IP) demonstrated ALF. 3.4. Illustrative case of ALF Case JM was a highly successful engineer who developed seizures in adulthood. He had a left temporal lobe tumor removed, but seizures recommenced six months after his operation. When assessed one year after his operation, he complained of poor memory, yet standard tests indicated no difficulties. It was only when his longer term memory was measured that a memory deficit became evident (see Fig. 1). Without longer term testing, his complaints would not have been substantiated or well understood. 3.5. Comparison of patients with ALF to those with memory impairment at 30 min Table 6 provides the details of the patients who showed memory impairment after a 30-min delay. As can be seen, there were several instances of patients failing on more than one test. However, a comparison of the cases in Tables 5 and 6 reveals that in this sample of patients with focal epilepsy, the proportion of patients showing memory impairment at 30 min (8/15) was almost the same as the percentage who showed ALF (7/15). Of course, by the definition used in this study, a subject cannot score poorly at 30 min and go on to display ALF on the same task. However, as can be gleaned from comparing Tables 5 with Table 6, only 3 of the 7 patients who showed ALF had any memory impairment detected at the 30-min delay interval; patients MJ, DV, IP, and PO showed only ALF. Contrary to our prediction, the likelihood of having a hippocampal lesion was not higher for the group with an early memory impairment (2/8) than for those with ALF (4/7). Overall, 12/15 patients showed some sort of memory impairment. 4. Discussion Accelerated long-term forgetting is an important type of memory disorder that can cause a good deal of distress and shows a strong association with everyday memory complaints [11]. Until recently, because of the limits of standard clinical testing, neuropsychologists have not had the tools to detect ALF. For this reason, we sought to provide norms for long-term memory assessment that could be used to diagnose this condition. This study has met its four aims. First, normative data on three clinically relevant tests were gathered from a control group of healthy subjects 18–60 years of age. Second, we found that scores from all three of these measures revealed patterns consistent with ALF. Across the three tests, almost half of the patients with focal epilepsy demonstrated this phenomenon, and, as expected, this was significantly higher than the rates seen in the NC group. Third, consideration of demographic and clinical variables yielded some insights into factors associated with ALF, namely, older age and higher estimated IQ. Fourth, in this small sample composed of patients with temporal focal epilepsy and extratemporal focal epilepsy, we found that rates of ALF were similar to rates of memory impairment at 30-min delay, and having a hippocampal lesion was not associated with either. Importantly, using these new norms, we identified four patients (27% of our sample) as having longer term memory deficits (i.e., ALF) but no detectable memory deficit after a 30-min delay.



5

Table 5 Characteristics of individual patients who showed ALF. Cases showing ALF on each testa

Site of epileptic focus

Side of epileptic focus

Presence of hippocampal lesion

Sex

Age

Education

IQ

Years of seizures

RAVLT PW JM DB JH

Temporal Temporal Temporal Temporal

Right Left Left Left

Yes Yes No No

M M F M

47 40 58 61

12 18 13 15

104 115 114 96

34 1 b1 49

Logical Memory IP JM

Extratemporal Temporal

Right Left

No Yes

M M

33 40

12 18

83 115

1 1

Aggie Figures PO SC

Temporal Temporal

Left Left

Yes Yes

M F

40 39

15 13

89 91

19 5

No NC subject showed ALF on more than one test. a The numbers of NC subjects who met these criteria for ALF were as follows: n = 3 (RAVLT); n = 2 (Logical Memory); and n = 5 (Aggie Figures).

4.1. Consideration of the normative values obtained

4.2. Detecting ALF in patients with focal epilepsy and the factors implicated

This is the first time that 30-min and 7-day norms have been provided for three memory tests that are frequently used (or have been recommended for use [27]) in epilepsy clinics. The age range sampled (i.e., 18–60) is appropriate for a high proportion of the patients seen in epilepsy clinics. Now, when patients present with memory complaints, the detection of ALF should be greatly facilitated. For the two verbal tests, the mean 30-min delayed recall scores obtained by our NC group were consistent with previously published age-based norms. Like before, differences in male and female verbal memory over short delays [21,29,43,45] were found. Interestingly, the sex effect was not seen in Percent Change over the 7-day delay for the verbal measures. Similarly, in the larger RAVLT data set collected by Geffen et al. [23], there was no clear difference in rate of memory decline between men and women tested at 30 min and 7 days. For the Aggie Figures test, the mean 30-min delay recall score (11.3) that we obtained was slightly lower than the mean scores (13–14) obtained by Majdan et al. [27] and Sziklas and Jones-Gotman [46] for larger groups of NC subjects. Because the mean scores are only provided graphically in those previous publications and no standard deviations are supplied, it is difficult to determine whether this represents a significant difference in performance. It may be worth noting that the groups in the previous publications had higher mean levels of education compared with the current sample, and we did find that Aggie Figures memory (at least at very long delays) was better in more highly educated subjects. Future research with this measure should continue to consider the possible impact of education level.

There are different ways to operationalize the definition of ALF. We have used a method that is fairly commonly employed [16,22,47]. We have chosen to use cutoff scores of 1.64 SD (5th percentile) for defining normal and abnormal performances. Applying the 1.64 SD cutoff scores resulted in 47% of patients and only 17% of NCs being identified as showing ALF. We note that when a less stringent cutoff (−1 SD) was applied to 30-min story recall and 2-week story recall [19], no difference in frequency of “ALF” was found between NCs and patients with TLE. We would argue that cutoffs at z b − 1 SD are too lenient, but, given that norms are now provided, clinicians can choose to use different cutoffs to define normal and impaired performances, depending on issues of specificity and sensitivity appropriate to their setting. One might also like to consider whether a patient shows a higherthan-expected rate of forgetting over longer intervals irrespective of their memory performance at initial testing (in other words, a high rate of decay in addition to a low score on earlier recall). This definition of ALF is most often applied when material is not initially learned to criterion (e.g., [7,19,48]). The Percent Change scores provided here and used in isolation could determine which patients show ALF according to this criterion. This would result in a higher number of subjects being classified as showing ALF. It is also possible that a full week's delay is not necessary to detect ALF. Recent findings for patients with epilepsy suggest significant loss by 24 h in those with hippocampal lesions [32,48] and possibly even more rapid deterioration for those with TEA [49,50]. We chose a 7-day delay because it has been noted that some patients with epilepsy show significant memory decline only over longer time periods [32,48]. Using the longer interval will ensure that there are fewer false negatives (i.e., reporting that patients do not have a memory disorder, when, in fact, they do). Jones-Gotman and colleagues stress the usefulness of learning tasks involving abstract line drawings for detecting deficits in patients with right temporal lesions [27,28]. We chose to include the Aggie Figures test as a visual memory measure because of its similarity to the RAVLT. We found that it was not difficult to administer or score. It did seem to measure something different from the two verbal memory measures as some patients demonstrated deficits only on the Aggie Figures Test; however, it was not uncommon for these cases to have a leftsided lesion. Thus, the usefulness of Aggie Figures in discriminating right-sided foci from left-sided foci in delayed recall was not particularly evident. It remains possible that left–right differences might have been found had we examined performance over the learning trials. With respect to epilepsy variables, we found that ALF was seen predominantly, but not exclusively, in patients with TLE. This is consistent with recent findings, which indicate that ALF can occur in patients with epileptic syndromes other than TLE [10]. A larger patient sample

Fig. 1. Case JM, a patient with ALF. His performances on the three memory measures are shown as standard scores based on the NC sample means. * indicates Percent Change score falling N1.64 SD from NC mean.


6


Table 6 Characteristics of patients who showed memory impairments at standard 30-min delayed recall. Cases showing recall deficit at 30 mina

Site of epileptic focus

Side of epileptic focus

Presence of hippocampal lesion

Age

Education

IQ

Years of seizures

RAVLT LC TV RR SC ZM

Extratemporal Extratemporal Temporal Temporal Temporal

Unknown Bilateral Left Left Right

No No No Yes Yes

32 31 26 39 45

10 11 15 13 15

100 85 90 91 96

11 1 1 5 17

Logical Memory JI LC TV ZM

Extratemporal Extratemporal Extratemporal Temporal

Right Unknown Bilateral Right

No No No Yes

31 32 31 45

11 10 11 15

91 100 85 96

16 11 1 17

Aggie Figures PW JI TV JH

Temporal Extratemporal Extratemporal Temporal

Right Right Bilateral Left

Yes No No No

47 31 31 61

12 11 11 15

104 91 85 96

34 16 1 49

Fourteen (23%) of the NC subjects showed impairments on at least one of the tests. a Using z b −1.64 as the cutoff score, the numbers of NC subjects who showed memory impairments at 30 min were as follows: n = 6 (RAVLT); n = 6 (Logical Memory); and n = 8 (Aggie Figures).

would help determine whether it is more common in those with TLE, as is suggested by the pattern in the present data and, in general, by the fact that ALF is usually reported in cases with TLE [33]. We saw no side-of-lesion effects for ALF, which has also been found in some previous studies [22,31]. Our patient sample was small (n = 15), but, consistent with previous research (e.g., [9,22,32,37]), we found no evidence that a hippocampal lesion was necessary for patients with epilepsy to show ALF. On the other hand, as predicted based on our review of the literature [33], older age and higher estimated IQ were associated with ALF, at least on the RAVLT. It must also be noted that our study also included few NC participants with b 12 years of education (n = 11), and, on at least one measure (Aggie Figures), education seemed to influence the forgetting rate (i.e., lower education associated with greater forgetting and also greater variability in scores). Hence, until a larger normative sample is obtained, the present cutoffs should be used with caution in patients with low levels of education. It has been suggested that epileptiform discharges or seizures during the delay interval might contribute to ALF [2,33,34], but these events were not recorded in the present study. In the future, the norms provided here could facilitate the conduction of regression analyses to determine the relative importance of these and other variables in a larger sample. In summary, this study determined norms that will be useful in clinical settings where patients with epilepsy have cognitive evaluations. Requesting recalls at 7-day delay on these commonly used standard measures will give neuropsychologists the ability to diagnose ALF. Sensitivity to this disorder is maximized by using multiple memory tests, as subjects rarely showed ALF on more than one of the three measures used here. Our results suggest that higher age and IQ are linked with ALF for a word list and that factors such as the presence of a hippocampal lesion, side of epilepsy focus, and duration of epilepsy are not predictive. However, the number of patients included here was small, and the findings with regard to factors contributing to ALF are tentative. Further studies, particularly those monitoring rates of epileptiform activity over the delay interval, are needed to determine which other factors contribute to ALF. Use of the norms provided will facilitate such endeavors. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yebeh.2015.01.024. Acknowledgments This research was supported by the ARC Centre of Excellence Grant CE110001021 in Cognition and Its Disorders.

Disclosure The authors have no conflicts of interest to disclose.

References [1] Butler CR, Graham KS, Hodges JR, Kapur N, Wardlaw JM, Zeman AZ. The syndrome of transient epileptic amnesia. Ann Neurol 2007;61(6):587–98. [2] Manes F, Graham KS, Zeman A, de Lujan Calcagno M, Hodges JR. Autobiographical amnesia and accelerated forgetting in transient epileptic amnesia. J Neurol Neurosurg Psychiatry 2005;76(10):1387–91. [3] Kapur N, Millar J, Colbourn C, Abbott P, Kennedy P, Docherty T. Very long-term amnesia in association with temporal lobe epilepsy: evidence for multiple-stage consolidation processes. Brain Cogn 1997;35:58–70. [4] Kapur N, Scholey K, Moore E, Barker S, Brice J, Thompson S, et al. Long-term retention deficits in two cases of disproportionate retrograde amnesia. J Cogn Neurosci 1996;8:416–35. [5] Mayes AR, Isaac CL, Holdstock JS, Cariga P, Gummer A, Roberts N. Long-term amnesia: a review and detailed illustrative case study. Cortex 2003;39:567–603. [6] Blake RV, Wroe SJ, Breen EK, McCarthy RA. Accelerated forgetting in patients with epilepsy: evidence for an impairment in memory consolidation. Brain J Neurol 2000;123(3):472–83. [7] Mameniskiene R, Jatuzis D, Kaubrys G, Budrys V. The decay of memory between delayed and long-term recall in patients with temporal lobe epilepsy. Epilepsy Behav 2006;8(1):278–88. [8] Muhlert N, Grunewald RA, Hunkin NM, Reuber M, Howell S, Reynders H, et al. Accelerated long-term forgetting in temporal lobe but not idiopathic generalised epilepsy. Neuropsychologia 2011;49(9):2417–26. [9] Butler CR, Bhaduri A, Acosta-Cabronero J, Nestor PJ, Kapur N, Graham KS, et al. Transient epileptic amnesia: regional brain atrophy and its relationship to memory deficits. Brain J Neurol 2009;132(2):357–68. [10] Gascoigne MB, Barton B, Webster R, Gill D, Antony J, Lah SS. Accelerated long-term forgetting in children with idiopathic generalized epilepsy. Epilepsia 2012;53(12): 2135–40. [11] Fitzgerald Z, Thayer Z, Mohamed A, Miller LA. Examining factors related to accelerated long-term forgetting in epilepsy using ambulatory EEG monitoring. Epilepsia 2013;54(5):819–27. [12] Helmstaedter C, Hauff M, Elger CE. Ecological validity of list-learning on a selective reminding procedure and self-reported memory in healthy individuals and those with temporal lobe epilepsy. J Clin Exp Neuropsychol 1998;20:365–75. [13] Butler CR, Zeman AZ. Recent insights into the impairment of memory in epilepsy: transient epileptic amnesia, accelerated long-term forgetting and remote memory impairment. Brain J Neurol 2008;131(9):2243–63. [14] Jansari AS, Davis K, McGibbon T, Firminger S, Kapur N. When “long-term memory” no longer means “forever”: analysis of accelerated long-term forgetting in a patient with temporal lobe epilepsy. Neuropsychologia 2010;48(6):1707–15. [15] Muhlert N, Milton F, Butler CR, Zeman AZ. Accelerated forgetting of real-life events in transient epileptic amnesia. Neuropsychologia 2010;48:3235–44. [16] Narayanan J, Duncan R, Greene J, Leach J-P, Razvi S, McLean J, et al. Accelerated longterm forgetting in temporal lobe epilepsy: verbal, nonverbal and autobiographical memory. Epilepsy Behav 2012;25(4):622–30. [17] Tramoni E, Felician O, Barbeau EJ, Guedj E, Guye M, Bartolomei F, et al. Long-term consolidation of declarative memory: insight from temporal lobe epilepsy. Brain 2011;134(Pt 3):816–31.


L.A. Miller et al. / Epilepsy & Behavior xxx (2015) xxx–xxx [18] Baddeley A, Rawlings B, Hayes A. Constrained prose recall and the assessment of long-term forgetting: the case of ageing and the Crimes Test. Memory 2013;22(8): 1052–9. [19] Bell B. WMS-III logical memory performance after a two-week delay in temporal lobe epilepsy and control groups. J Clin Exp Neuropsychol 2006;28:1435–43. [20] Kemp S, Illman NA, Moulin CJ, Baddeley AD. Accelerated long-term forgetting (ALF) and transient epileptic amnesia (TEA): two cases of epilepsy-related memory disorder. Epilepsy Behav 2012;24(3):382–8. [21] Chapman RM, Mapstone M, Gardner MN, Sandoval TC, McCrary JW, Guilily MD, et al. Women have farther to fall: gender differences between normal elderly and Alzheimer's disease in verbal memory engender better detection of AD in women. J Int Neuropsychol Soc 2011;17(4):654–62. [22] Wilkinson H, Holdstock JS, Baker G, Herbert A, Clague F, Downes JJ. Long-term accelerated forgetting of verbal and non-verbal information in temporal lobe epilepsy. Cortex 2011;48(3):317–32. [23] Geffen GM, Geffen L, Bishop K, Manning L. Extended delayed recall of AVLT word lists: effects of age and sex on adult performance. Aust J Psychol 1997;49(2):78–84. [24] Elliott G, Isaac CL, Muhlert N. Measuring forgetting: a critical review of accelerated long-term forgetting studies. Cortex 2014;54:16–32. [25] Rausch R, Le M-T, Langfitt JT, editors. Neuropsychological evaluation — adults. Philadelphia: Lippincott-Raven; 1997. [26] Jones-Gotman M. Localization of lesions by neuropsychological testing. Epilepsia 1991;32(Suppl. 5):S41–52. [27] Majdan A, Sziklas V, Jones-Gotman M. Performance of healthy subjects and patients with resection from the anterior temporal lobe on matched tests of verbal and visuoperceptual learning. J Clin Exp Neuropsychol 1996;18(3):416–30. [28] Jones-Gotman M. Psychological evaluation for epilepsy surgery. In: Shorvon S, Fish F, Thomas D, editors. The treatment of epilepsy. Oxford: Blackwell Science Ltd; 1996. p. 621–30. [29] Zelinski EM, Gilewski MJ, Schaie KW. Individual-differences in cross-sectional and 3year longitudinal memory performance across the adult life-span. Psychol Aging 1993;8(2):176–86. [30] Eals M, Silverman I. The hunter–gatherer theory of spatial sex-differences — proximate factors mediating the female advantage in recall of object arrays. Ethol Sociobiol 1994; 15(2):95–105. [31] Evans SJ, Elliott G, Reynders H, Isaac CL. Can temporal lobe epilepsy surgery ameliorate accelerated long-term forgetting? Neuropsychologia 2014;53:64–74. [32] Lah S, Mohamed A, Thayer Z, Miller L, Diamond K. Accelerated long-term forgetting of verbal information in unilateral temporal lobe epilepsy: is it related to structural hippocampal abnormalities and/or incomplete learning? J Clin Exp Neuropsychol 2014;36(2):158–69. [33] Fitzgerald Z, Mohamed A, Ricci M, Thayer Z, Miller L. Accelerated long-term forgetting: a newly identified memory impairment in epilepsy. J Clin Neurosci 2013;20(11): 1486–91.

7

[34] O'Connor M, Sieggreen MA, Ahern G, Schomer D, Mesulam M. Accelerated forgetting in association with temporal lobe epilepsy and paraneoplastic encephalitis. Brain Cogn 1997;35(1):71–84. [35] Holdstock J, Mayes A, Isaac C, Gong Q, Roberts N. Differential involvement of the hippocampus and temporal lobe cortices in rapid and slow learning of new semantic information. Neuropsychologia 2002;40:748–68. [36] Gallassi R, Sambati L, Poda R, Maserati MS, Oppi F, Giulioni M, et al. Accelerated longterm forgetting in temporal lobe epilepsy: evidence of improvement after left temporal pole lobectomy. Epilepsy Behav 2011;22(4):793–5. [37] Bengner T, Malina T, Lindenau M, Voges B, Goebell E, Stodieck S. Face memory in MRI-positive and MRI-negative temporal lobe epilepsy. Epilepsia 2006;47(11): 1904–14. [38] Milner B. Disorders of learning and memory after temporal lobe lesions in man. Clin Neurosurg 1972;19:421–46. [39] Baxendale SA, van Paesschen W, Thompson PJ, Connelly A, Duncan JS, Harkness WF, et al. The relationship between quantitative MRI and neuropsychological functioning in temporal lobe epilepsy. Epilepsia 1998;39(2):158–66. [40] Helmstaedter C, Pohl C, Hufnagel A, Elger CE. Visual learning-deficits in nonresected patients with right temporal-lobe epilepsy. Cortex 1991;27(4):547–55. [41] Hermann BP, Wyler AR, Richey ET, Rea JM. Memory function and verbal-learning ability in patients with complex partial seizures of temporal-lobe origin. Epilepsia 1987;28(5):547–54. [42] Rey A. L'examen psychologique dans les cas d'encephalopathie traumatique. Arch Psychol 1941;28:21. [43] PearsonEducation. Wechsler Memory Scale. Clinical features of the new edition. 4th ed. 2008. [44] Wechsler D. Test of premorbid functioning. San Antonio: The Psychological Corporation; 2009. [45] Strauss E, Sherman EMS, Spreen O. A compendium of neuropsychological tests: administration, norms, and commentary. 3rd ed. New York, NY: Oxford University Press; 2006. [46] Sziklas V, Jones-Gotman M. RAVLT and nonverbal analog: French forms and clinical findings. Can J Neurol Sci 2008;35:323–30. [47] Butler C, Kapur N, Zeman A, Weller R, Connelly A. Epilepsy-related long-term amnesia: anatomical perspectives. Neuropsychologia 2012;50(13):2973–80. [48] Ricci M, Mohamed A, Savage G, Boserio J, Miller LA. The impact of epileptiform abnormalities and hippocampal lesions on retention of recent autobiographical experiences: adding insult to injury? Neuropsychologia 2015;66:259–66. [49] Hoefeijzers S, Dewar M, Della Sala S, Zeman A, Butler C. Accelerated long-term forgetting in transient epileptic amnesia: an acquisition or consolidation deficit? Neuropsychologia 2013;51(8):1549–55. [50] Atherton KE, Nobre AC, Zeman AZ, Butler CR. Sleep-dependent memory consolidation and accelerated forgetting. Cortex 2014;54:92–105.


New dogs, old tricks.

On Teaching Old Dogs New Tricks.

tRNAs: new tricks from old dogs.

A medical homily: old dogs do teach new tricks.

Old Dogs Learning New Tricks: Neuroplasticity Beyond the Juvenile Period.

Myosin light chains: Teaching old dogs new tricks.

Evidence-based maternity care: can new dogs learn old tricks?

Memory avoidance by older adults: When `old dogs' won't perform their `new tricks'.

Arginase: an old enzyme with new tricks.

New tricks for old dogs: countering antibiotic resistance in tuberculosis with host-directed therapeutics.

Pristinamycin: old drug, new tricks?

New (or not-so-new) tricks for old dogs: ultrasound imaging in anatomy laboratories.

"Old Dogs with New Tricks": exploiting alternative mechanisms of action and new drug design strategies for clinically validated HIV targets.

NSAIDs: learning new tricks from old drugs.

Baculovirus expression: old dog, new tricks.

DNA methylation: old dog, new tricks?

Antimicrobials: New tricks for old drugs.

New tricks for an old workhorse.

Water-pipe smoking and albuminuria: new dog with old tricks.

Advanced endoscopic training: teaching us older dogs some new tricks.

Old drugs, new tricks for triple-negative breast cancer.

Pinaverium in Irritable Bowel Syndrome: Old Drug, New Tricks?

Old dog, new tricks: Arf1 required for mitochondria homeostasis.

6,6-Dicyanopentafulvenes: teaching an old dog new tricks.