INTRODUCTION
Huntington’s disease (HD) is a progressive neurodegenerative disorder caused by an expansion of CAG repeats on the short arm of chromosome 4 (Huntington’s Disease Collaborative Group, 1993). The disease is characterized by motor abnormalities, psychiatric disturbance, and cognitive deterioration. A clinical diagnosis of HD is based on the presence of motor symptoms. However, several studies have shown cognitive and psychiatric changes in individuals who carry the HD expansion but do not meet criteria for a clinical diagnosis of HD (referred to as preclinical HD or pre-HD).
Several neuropsychological studies have detected cognitive changes prior to the clinical manifestation of HD, while other studies have not found subtle cognitive deficits in pre-HD (for review, see Georgiou-Karistianis, Churchyard, Chiu, & Bradshaw, Reference Georgiou-Karistianis, Churchyard, Chiu and Bradshaw2002). Many of the neuropsychological deficits reported in pre-HD fall in the cognitive domains of executive function, memory, and psychomotor speed (Hahn-Barma et al., Reference Hahn-Barma, Deweer, Durr, Dode, Feingold and Pillon1998; Kirkwood et al., Reference Kirkwood, Siemers, Stout, Hodes, Conneally and Christian1999, Reference Kirkwood, Siemers, Hodes, Conneally, Christian and Foroud2000; Lemiere, Decruyenaere, Evers-Kiebooms, Vandenbussche, & Dom, Reference Lemiere, Decruyenaere, Evers-Kiebooms, Vandenbussche and Dom2002; Paulsen et al., Reference Paulsen, Zhao, Stout, Brinkman, Guttman and Ross2001; Pirogovsky et al., Reference Pirogovsky, Gilbert, Jacobson, Peavy, Wetter and Goldstein2007; Snowden, Crawford, Thompson, & Neary, Reference Snowden, Crawford, Thompson and Neary2002; Solomon et al., Reference Solomon, Stout, Johnson, Langbehn, Aylward and Brandt2007; Wahlin, Lundin, & Dear, Reference Wahlin, Lundin and Dear2007). Cognitive deficits in pre-HD are thought to stem from striatal atrophy and consequent dysfunction in frontostriatal pathways. Neuroimaging studies have reported changes in the striatum in pre-HD, with some studies showing volumetric reductions in the striatum beginning as early as 7 to 15 years prior to clinical manifestation of HD (Aylward et al., Reference Aylward, Brandt, Codori, Mangus, Barta and Harris1994, Reference Aylward, Codori, Barta, Pearlson, Harris and Brandt1996, Reference Aylward, Codori, Rosenblatt, Sherr, Brandt and Stine2000; Paulsen et al., Reference Paulsen, Hayden, Stout, Langbehn, Aylward and Ross2006). In addition, studies have reported correlations between striatal atrophy and slower psychomotor speed as well as impaired verbal learning in pre-HD (Campodonico et al., Reference Campodonico, Aylward, Codori, Young, Krafft and Magdaliniski1998; Solomon et al., Reference Solomon, Stout, Johnson, Langbehn, Aylward and Brandt2007). Studies also report reductions in white matter volume (Ciarmiello et al., Reference Ciarmello, Cannella, Lastoria, Simonelli, Frati and Rubinsztein2006), metabolic changes (Ciarmiello et al., Reference Ciarmello, Cannella, Lastoria, Simonelli, Frati and Rubinsztein2006; Gomez-Anson et al., Reference Gomez-Anson, Alegret, Munoz, Sainz, Monte and Tolosa2007), and functional changes (Reading et al., Reference Reading, Dziorny, Peroutka, Schreiber, Gourley and Yallapragada2004; Wolf et al., Reference Wolf, Vasic, Schonfeldt-Lecuona, Landwehrmeyer and Ecker2007; Zimbelman et al., Reference Zimbelman, Paulsen, Mikos, Reynolds, Hoffmann and Rao2007) in the frontal cortex of pre-HD. Thus, neuroimaging studies support the notion that cognitive dysfunction in pre-HD is associated with changes in frontostriatal circuitry.
Temporal order memory is defined as memory for the order in which items or events have been experienced. Evidence from humans (Kesner, Hopkins, & Fineman, Reference Kesner, Hopkins and Fineman1994; Kopelman, Stanhope, & Kingsley, Reference Kopelman, Stanhope and Kingsley1997; McAndrews and Milner, Reference McAndrews and Milner1991; Milner, Reference Milner1971; Milner, Corsi, & Leonard, Reference Milner, Corsi and Leonard1991; Shimamura, Janowksy, & Squire, Reference Shimamura, Janowsky and Squire1990), non-human primates (Inoue and Mikami, Reference Inoue and Mikami2006; Ninokora, Mushiake, & Tanji, Reference Ninokura, Mushiake and Tanji2004), and rats (Chiba Kesner, & Gibson, Reference Chiba, Kesner and Gibson1997; Hanneson, Vacca, Howland, & Phillips, Reference Hanneson, Vacca, Howland and Phillips2004) with frontal lobe lesions suggests that the frontal cortex plays an important role in processing temporal order information. The frontal cortex may support processes that temporally organize sequences of events (Fuster, Reference Fuster2001). Because pre-HD demonstrates dysfunction in frontostriatal circuits, temporal order memory may be sensitive to neuropathological dysfunction in pre-HD. A few studies have shown evidence of impairments in temporal perception (e.g., time discrimination) and motor reproduction (e.g., finger tapping) in pre-HD (Hinton et al., Reference Hinton, Paulsen, Hoffmann, Reynolds, Zimbelman and Rao2007; Paulsen et al., Reference Paulsen, Zimbelman, Hinton, Langbehn, Leveroni and Benjamin2004), suggesting a possible general deficit in temporal processing. Furthermore, pre-HD demonstrate deficits in motor sequence learning (Feigin et al., Reference Feigin, Ghilardi, Huang, Ma, Carbon and Guttman2006; Ghilardi et al., Reference Ghilardi, Silvestri, Feigin, Mattis, Zgaljardic and Moisello2008) and picture sequencing (Foroud et al., Reference Foroud, Seimers, Kleindorfer, Bill, Hodes and Norton1995; Snowden et al., Reference Snowden, Crawford, Thompson and Neary2002) in pre-HD, suggesting a potential deficit in temporal sequencing and organization. However, no study to date has investigated temporal order memory in pre-HD.
Neuropsychological studies in pre-HD use equations with relevant predictors, such as CAG repeat length, to estimate the number of years to clinical diagnosis of HD. By examining relationships between estimated years to HD onset and cognitive deficits, cross-sectional studies can predict how far in advance different cognitive deficits arise as well as the progression of cognitive impairments in pre-HD. The current study used an equation developed by Aylward and colleagues (Reference Aylward, Sparks, Field, Yallapragada, Shpritz and Rosenblatt2004) that predicts years to diagnosis of manifest HD with CAG repeat length, participant’s age, and parent’s age when diagnosed with manifest HD. The purpose of the current study was to use a novel experimental task to examine temporal order memory in pre-HD close to estimated disease diagnosis compared with pre-HD further from estimated disease diagnosis and normal controls. The task used parametric manipulations of the temporal metric by systematically changing the number of items between two choice items in a temporal sequence (i.e., temporal separation lag). Research demonstrates that temporal order memory is better for items occurring further apart in a temporal sequence compared with items that are temporally proximal (Kesner & Hopkins, Reference Kesner and Hopkins2006). This is referred to as the temporal separation effect, and it is hypothesized to occur because there is more interference for temporally proximal stimuli than temporally distant stimuli (Gilbert, Kesner, & Lee, Reference Gilbert, Kesner and Lee2001). Therefore, varying temporal interference by manipulating the temporal lag allows for the examination of deficits that range from subtle to severe using a single paradigm. Examining temporal interference in temporal order memory in pre-HD may elucidate a fundamental but virtually unexamined processing deficit and could potentially inform behavioral interventions that help structure daily living tasks to mitigate interference in the temporal domain.
METHODS
Participants
Pre-HD (n = 18) participants were recruited from the HD Clinical Research Program at the University of California, San Diego. Normal control participants (n = 18) were recruited from the San Diego Community. Pre-HD diagnosis was based on number of CAG repeats (≥39) in individuals without manifest HD motor symptoms. The motor exam subsection of the Unified Huntington’s Disease Rating Scale (UHDRS; Huntington Study Group, 1996) was used to screen for symptoms of manifest HD in pre-HD. The UHDRS was administered by a senior staff neurologist or an experienced HD clinician who was trained and certified to complete the evaluation. The mean total motor score for the total sample was 2.3 (SE = .62) out of a possible 124. Based on the UHDRS motor exam, the examiner assigned a diagnostic confidence rating representing the examiner’s confidence that the presence of abnormalities were a manifestation of HD. A confidence score of 4 is considered the point at which HD diagnosis is made. The ratings are defined as: 0 = normal (no abnormalities), 1 = non-specific motor abnormalities (<50% confidence that the participant has sufficient motor abnormalities to warrant a diagnosis of HD), 2 = motor abnormalities that may be signs of HD (50–89% confidence), 3 = motor signs that are likely signs of HD (90–98% confidence), 4 = motor abnormalities that are unequivocal signs of HD (≥99% confidence). In the pre-HD group, 1 participant received a score of 2, 6 participants received a score of 1, and 11 participants received a score of 0. Therefore, none of the participants met criteria for a diagnosis of manifest HD. Exclusion criteria for the pre-HD group included: clinical evidence of an active psychiatric illness (i.e., within the last 6 months); history of other neurological conditions, such as seizures or head injury; and history of alcohol or other substance abuse. For the normal control group, exclusion criteria included: clinical evidence of an active psychiatric illness (i.e., within the last 6 months); history of a neurological condition, such as seizures, head injury, or neurodegenerative disease; and history of alcohol or other substance abuse.
The predicted age of HD diagnosis for the present sample was calculated using the method described by Aylward and colleagues (Reference Aylward, Sparks, Field, Yallapragada, Shpritz and Rosenblatt2004), which uses CAG repeat length and parents’ age of diagnosis: [predicted age of diagnosis = (−.81 × repeat length) + (.51 × parental age at diagnosis) + 54.87]. Predicted years to HD diagnosis was calculated by subtracting participants’ age on the day of testing from their predicted age of diagnosis. The mean predicted years to diagnosis for the pre-HD group was 5.2 years (SE = 1.72), ranging from −3.4 to 17.8 years. A negative predicted age of diagnosis suggests that an individual is estimated to already have been diagnosed with manifest HD. The pre-HD group was divided based on the mean predicted years to diagnosis. Thus, the groups consisted of those who were less than 5.2 years to predicted diagnosis (pre-HD near, n = 11) and those who were more than 5.2 years to predicted diagnosis (pre-HD far, n = 7). The mean estimated years to diagnosis were for the pre-HD near group, −.04 (SE = .72) and for the pre-HD far group, 13.3 (SE = 1.6). All participants who received a diagnostic confidence rating of greater than 0 were in the pre-HD near group. The mean total UHDRS score for the HD near and HD far groups was 3.6 (SE = 1.1) and 1.4 (SE = .48), respectively.
Participants were administered the Dementia Rating Scale (DRS; Mattis, Reference Mattis, Bellack and Katsau1976) to screen for dementia. A one-way analysis of variance (ANOVA) did not detect any differences among the three groups in DRS scores, F(2,33) = 2.8; p = .08. The demographic and genetic characteristics as well as performance on the DRS of all groups are provided in Table 1. A one-way ANOVA with group (pre-HD near, pre-HD far, normal control) as the between-group factor revealed a significant effect of age, F(2,33) = 4.2; p < .05. Tukey comparison tests revealed that the pre-HD near group was significantly older relative to the pre-HD far group (p < .05). There were no significant differences detected between the pre-HD far group and normal controls. A one-way ANOVA did not detect any group differences among the three groups in level of education, F(2,33) = .93; p = .40.
Table 1. Mean (SE) DRS scores as well as demographic and genetic information of pre-HD far, pre-HD near, and normal control participants
Note
DRS = Dementia Rating Scale; HD = Huntington’s disease.
All participants were provided written informed consent documents according to the Declaration of Helsinki and approved by both San Diego State University and the University of California, San Diego.
Temporal Order Memory Task
The temporal order task used in the current study was modeled after previously published paradigms (Hopkins, Kesner, & Goldstein, Reference Hopkins, Kesner and Goldstein1995; Hopkins, Waldram, & Kesner, Reference Hopkins, Waldram and Kesner2004; Kesner & Hopkins, Reference Kesner and Hopkins2006). The participant was seated in a chair approximately 60 cm from a 59-cm (diagonal) computer monitor. At the beginning of each trial, the participant was prompted to focus on the monitor where a computerized radial eight-arm maze was presented. The computerized eight-arm maze consisted of eight arms extending from a center like spokes on a wheel. The eight-arm maze appeared on the computer screen with a diameter of approximately 30 cm. The participant was told that a circle would appear at the end of each arm one at a time in a random sequence. The experimenter instructed the participant to remember the sequence in which the circles were presented on the arms.
Each trial consisted of a sample phase followed by a choice phase. On the sample phase, a gray circle (3-cm diameter) appeared at the end of a randomly selected arm. The circle appeared for 2 s, and then the entire display was masked for 2 s by a gray mask to eliminate after image effects. Then another circle appeared at the end of a different randomly selected arm for 2 s followed by a 2-s mask. This continued until a gray circle had been presented at the end of each of the eight arms once in a random sequence that varied on each trial (Figure 1a). On the choice phase, the participant was presented simultaneously for 5 s with two circles, one at the end of one of the study phase arms and the other at the end of another study phase arm. The participant was asked to indicate which arm appeared earlier in the sequence.
Fig. 1. A schematic of a random sample phase temporal sequence showing locations of the 1st through the 8th arms presented in a random sequence (A) and a choice phase (B) consisting a 6 temporal separation lag trial, a 2 temporal separation lag trial, and a 0 temporal separation lag trial.
Temporal separations of 0, 2, 4, and 6 lags were randomly selected for each choice phase. The temporal separation lags represented the number of arms that occurred in the sample phase sequence between the two arms presented simultaneously in the choice phase (Figure 1b). For example, on a 6 temporal separation lag, the participant was presented with two arms on the choice phase that occurred with six arms between them on the sample phase sequence (e.g., 1st circle presented vs. 8th circle presented). On a 2 temporal separation lag, two arms were presented on the choice phase that occurred with two arms between them on the sample phase sequence (e.g., 2nd circle presented vs. 5th circle presented). Following each sample phase sequence, three choice phases were conducted. Each of the three choice phase trials involved three of the four temporal separation lags (e.g., 0 lag, 2 lag, and 6 lag). A total of 16 different sample phase sequences were presented with three choice phases for each sequence. As a result, there were 12 choice phase trials for each of the four temporal separations. Each participant was administered the same 16 sample phases. A 15-s intertrial interval was implemented between each trial. The responses of the participant were recorded by the experimenter. The raw data for each of the four temporal separations were converted into percent correct score for analysis.
Neuropsychological Measures
Participants were administered the Trail Making Test (TMT), Color-Word Interference Test (CWIT), and Verbal Fluency subtests of the Delis Kaplan Executive Function System (D-KEFS; Delis, Kaplan, & Kramer, Reference Delis, Kaplan and Kramer2001); the Hopkins Verbal Learning Test-Revised (HVLT-R; Brandt & Benedict, 2001); the Verbal Paired Associates Subtest from the Wechsler Memory Scale-Third Edition (WMS-III; Wechsler, Reference Wechsler2002); and the Symbol Digit Modalities Test (Smith, Reference Smith1973) to examine cognitive profiles of executive function, memory, and psychomotor speed.
RESULTS
Neuropsychological Measures
One-way ANOVAs with group (pre-HD far, pre-HD near, normal control) as the independent variable were used to analyze the neuropsychological data. The results of the analysis are provided in Table 2. Analyses showed significant main effects of group on the HVLT-R total immediate recall index [F(1,33) = 6.6; p < .01; η 2 = .29], HVLT-R recognition discriminability index [F(1,33) = 12.9; p < .001; η2 = .45], the inhibition/switching condition of the CWIT [F (1,33) = 3.6; p < .05; η2 = .19], and the visual scanning condition of the TMT [F (1,33) = 4.1; p < .05; η2 = .20]. Tukey post hoc comparison tests of the significant effects showed performance on the total immediate recall index of the HVLT-R was significantly impaired in pre-HD near compared with the pre-HD far (p < .05) and normal control (p < .01) groups. Analyses also found that the recognition discriminability index of the HVLT-R was significantly lower (p < .01) in pre-HD near compared with pre-HD far and normal controls. In addition, the visual scanning condition of the Trail Making Test and the inhibition/switching condition of the Color Word Interference Test were significantly impaired (p < .05) in the pre-HD near group relative to the controls but not compared with the pre-HD far group. Statistical analyses did not detect significant between-group differences on scores from any other neuropsychological tests.
Table 2. Mean (SE) performance of pre-HD far, pre-HD near, and normal controls on the standardized neuropsychological tests
Note
Levels of significance for comparisons between pre-HD near and pre-HD far are shown in the pre-HD far column and for comparisons between pre-HD near and controls are shown in the controls column. Raw scores are reported. HD = Huntington’s disease; TMT = Trail Making Test, CWIT = Color Word Interference Test, VF = Verbal Fluency, HVLT = Hopkins Verbal Learning Test – Revised, VPA = Verbal Paired Associates.
* p < .05.
** p < .01.
Temporal Order Memory Task
To account for group differences in age, a 3 × 4 analysis of covariance (ANCOVA) with group (pre-HD far, pre-HD near, normal control) as a between-group factor, temporal separation lag (0, 2, 4, 6) as a within-group factor, and age as a covariate was used to analyze the data. The main effect of age [F(1,32) = 1.9; p = .18] and the age × lag interaction [F(3,96) = .22; p = .88] were not statistically significant. Therefore, a 3 × 4 ANOVA that excluded the covariate of age was used to analyze the data and the overall findings did not differ from the ANCOVA. Mauchly’s test of sphericity indicated that the assumption of sphericity was violated, χ2(5) = 12.17; p < .05. Thus, degrees of freedom were corrected using Huynh-Feldt estimates of sphericity (ε = .89). The analysis showed a significant main effect of group, a main effect of lag, and a group × lag interaction. The results of the analysis are provided in Table 3. A Newman-Keuls post hoc comparison test of the statistically significant group × lag interaction showed that performance on the 0, 2, and 4 temporal separation lags was significantly lower in the pre-HD near group (p < .05) relative to both the pre-HD far and normal control groups (Figure 2). However, no significant differences were detected between the pre-HD near, pre-HD far, and normal control groups on the 6 temporal separation lag. In addition, no significant differences were detected between the pre-HD far group and normal controls on the 0, 2, 4, or 6 temporal separation lags.
Fig. 2. Mean percent correct performance (± standard error) of participants ≤ 5 years to estimated age of diagnosis (pre-HD near), participants > 5 years to estimated age of diagnosis (pre-HD far), and normal controls on the 0, 2, 4, and 6 temporal separation lag trials of the temporal order memory task.
Table 3. Results from a 3 × 4 ANOVA with group (pre-HD near, pre-HD far, normal controls) as the between groups factor and lag (0, 2, 4, 6) as the within groups factor
Note
ANOVA = analysis of variance; HD = Huntington’s disease.
Pearson r correlational analyses were used to examine the relationship between years to estimated diagnosis in all pre-HD participants (i.e., pre-HD near and pre-HD far groups combined) and mean percent correct performance on each temporal separation lag of the temporal order memory task. Results found significant correlations between estimated years to diagnosis in pre-HD and performance on the 2 temporal separation lag (r = .59; p < .05) and 4 temporal separation lag (r = .49; p < .05), but not the 0 temporal separation lag (r = .13; p = .60) or the 6 temporal separation lag (r = .15; p = .55).
Temporal order memory may account for between-group variability on the neuropsychological measures. To examine this possibility, further statistical analyses were conducted using neuropsychological scores that showed significant between-group differences (i.e., HVLT total immediate recall, HVLT recognition discriminability, inhibition/switching condition of the CWIT, visual scanning condition of the TMT). An average score on the temporal order memory task was computed by averaging the lags that demonstrated statistically significant between-group differences, including the 0, 2, and 4 lags. Separate univariate ANCOVAs with the average score as a covariate, group as the between-group variable, and HVLT total immediate recall, HVLT recognition discriminability, CWIT inhibition/switching, and TMT visual scanning as dependent variables were conducted. Results found that even when controlling for temporal order memory performance, the main effect of group on HVLT total immediate recall [F(1,32) = 8.1; p < .01] and recognition discriminability [F(1,32) = 3.7; p < .05] remained significant. However, the main effects of group on CWIT inhibition/switching [F(1,32) = 1.4; p = .27] and TMT visual scanning [F(1,32) = 2.9; p = .12] were no longer statistically significant.
DISCUSSION
The present study examined the effects of temporal interference on temporal order memory in pre-HD. The results of the study found that temporal order memory was impaired in the pre-HD near group relative to the pre-HD far and normal control groups; however, performance differed as a function of increasing temporal separation lag. Specifically, the pre-HD near group demonstrated impaired performance on proximal separation lags (0, 2, 4 lags); however, performance on the most distal lag (6 lag) was intact. The pre-HD far group matched the performance of normal controls on all temporal separation lags of the temporal order task. The results suggest that temporal order memory deficits are detectable in pre-HD up to 5 years before HD diagnosis when there is greater temporal interference.
The results also showed significant correlations between estimated years to HD diagnosis in the pre-HD group (i.e., pre-HD far and pre-HD near combined) and performance on the 2 and 4 temporal separation lags, but the correlations between estimated years to onset and performance on the 0 and 6 lags were not significant. A potential explanation for the lack of a statistically significant correlation between years to onset and the 0 lag may be that the pre-HD far group showed higher variability in performance on the 0 lag (see Figure 2). Thus, some pre-HD far were performing much better than others who were far from onset, precluding a relationship between estimated years to onset and 0 lag performance. Given the high performance of the pre-HD near and pre-HD far groups on the 6 lag, the non-significant correlation between estimated years to onset and the 6 lag was expected. These patterns of correlations suggest that estimated proximity to HD diagnosis is associated with increasingly poorer performance on the 2 and 4 temporal lags of the temporal order memory task.
Temporal order memory impairments in pre-HD may be related to neuropathological changes in the striatum and resultant disruption in frontostriatal circuits. Neuroimaging studies (e.g., structural, functional, and diffusion tensor imaging studies) have reported dysfunction in the frontal cortex, striatum, and other components of frontostriatal circuitry in pre-HD (Aylward et al., Reference Aylward, Brandt, Codori, Mangus, Barta and Harris1994; Ciarmiello et al., Reference Ciarmello, Cannella, Lastoria, Simonelli, Frati and Rubinsztein2006; Gomez-Anson et al., Reference Gomez-Anson, Alegret, Munoz, Sainz, Monte and Tolosa2007; Paulsen et al., Reference Paulsen, Zimbelman, Hinton, Langbehn, Leveroni and Benjamin2004; Reference Paulsen, Hayden, Stout, Langbehn, Aylward and Ross2006; Reading et al., Reference Reading, Dziorny, Peroutka, Schreiber, Gourley and Yallapragada2004, Reference Reading, Yassa, Bakker, Dziorny, Gourley and Yallapragada2005; Rosas et al., Reference Rosas, Tuch, Hevelone, Zaleta, Vangel and Hersch2006; Thieben et al., Reference Thieben, Duggins, Good, Gomes, Mahant and Richards2002; Wolf et al., Reference Wolf, Vasic, Schonfeldt-Lecuona, Landwehrmeyer and Ecker2007; Zimbelman et al., Reference Zimbelman, Paulsen, Mikos, Reynolds, Hoffmann and Rao2007). As mentioned previously, there is converging evidence from both human and animal research demonstrating the role of the frontal cortex in memory for temporal order (Chiba et al., Reference Chiba, Kesner and Gibson1997; Kesner et al., Reference Kesner, Hopkins and Fineman1994; Milner et al., Reference Milner, Corsi and Leonard1991; Shimamura et al., Reference Shimamura, Janowsky and Squire1990). In addition, neuroimaging studies in healthy adults have shown that the prefrontal cortex is involved in temporal order memory (Cabeza et al., Reference Cabeza, Mangels, Nyberg, Habib, Houle and McIntosh1997; Knutson, Wood, & Grufman, Reference Knutson, Wood and Grafman2004; Suzuki et al., Reference Suzuki, Fujii, Tsukiura, Okuda, Umetsu and Nagasaka2002). Therefore, temporal order memory may be related to frontostriatal neuropathology in pre-HD.
Fuster (Reference Fuster2001) purports that a key function of the prefrontal cortex is the temporal integration of information for the attainment of prospective behavioral goals. The prefrontal cortex is thought to mediate contingencies across time, which is a fundamental aspect of the temporal organization of events (Fuster, Reference Fuster2001). Thus, the frontal cortex may support processes that organize sequences of events. Disruption in these processes may lead to impairments in cognitive domains that rely on goal directed behavior, such as executive functions. A prominent feature of HD is executive dysfunction (Brandt & Butters, Reference Brandt, Butters, Grant and Adams1997), and there is growing evidence to suggest that executive functioning deficits occur prior to disease diagnosis (Farrow et al., Reference Farrow, Churchyard, Chua, Bradshaw, Chiu and Georgiou-Karistianis2007; Kirkwood et al., Reference Kirkwood, Siemers, Hodes, Conneally, Christian and Foroud2000; Lawrence et al., Reference Lawrence, Hodges, Rosser, Kershaw, French-Constant and Rubinsztein1998; Lemiere et al., Reference Lemiere, Decruyenaere, Evers-Kiebooms, Vandenbussche and Dom2002; Paulsen et al., Reference Paulsen, Zhao, Stout, Brinkman, Guttman and Ross2001; Wahlin et al., Reference Wahlin, Lundin and Dear2007). Thus, deficits in temporal order memory in pre-HD may result in impairments in executive function, a cognitive domain that is essential to the execution of activities of daily living tasks. In the current study, the pre-HD near group was impaired relative to normal controls on one of the executive function measures, the inhibition/switching condition of the CWIT. Interestingly, this difference between pre-HD near relative to controls did not remain statistically significant after controlling for differences in temporal order memory performance. This suggests that impairments on this executive function index in pre-HD could be explained by between-group variability in temporal order memory. It is important to note that the pre-HD near group was not impaired relative to the other groups on any of the other standardized executive function measures. One possibility for lack of differences between the pre-HD near group and the other groups may be that these standardized executive function measures tap into different aspects of executive function than the current temporal order memory task.
In the current study, the pre-HD near group was impaired relative to the pre-HD far and normal control groups on indices of the HVLT-R including total immediate recall and recognition discriminability. However, no differences were detected on delayed recall or retention. These findings are in accordance with previous studies demonstrating impaired verbal episodic memory in pre-HD (Lemiere et al., Reference Lemiere, Decruyenaere, Evers-Kiebooms, Vandenbussche and Dom2002; Lundervold & Reinvang, Reference Lundervold and Reinvang1995; Solomon et al., Reference Solomon, Stout, Johnson, Langbehn, Aylward and Brandt2007) but are in contrast to other studies that did not detect verbal episodic memory deficits in pre-HD (Brandt, Shpritz, Codori, Margolis, & Rosenblatt, Reference Brandt, Shpritz, Codori, Margolis and Rosenblatt2002). For example, Solomon and colleagues (Reference Solomon, Stout, Johnson, Langbehn, Aylward and Brandt2007) found that probability of HD diagnosis in 5 years was significantly associated with all HVLT-R measures except for retention. Episodic memory impairment in individuals diagnosed with HD is thought to be a result of a general inability to initiate systematic retrieval of stored information, as opposed to a “cortical” dementia profile associated with deficient encoding and consolidation (Salmon & Filoteo, Reference Salmon and Filoteo2007). This memory profile is thought to be a result of frontostriatal dysfunction that leads to impairments in strategy or organization of information during retrieval. For example, HD patients show impaired free recall with relatively intact recognition memory, suggesting an impairment in the strategic aspects of retrieval. The findings from the current study are in general agreement with this pattern of memory deficits in pre-HD, with the exception that pre-HD were impaired in recognition discriminability and were not impaired in delayed recall.
Temporal order memory also may contribute to the formation and retrieval of episodic memories. For example, remembering a gathering with friends involves recalling a series of linked events not only in a spatial context or “where” (e.g., in a restaurant), but also a temporal context or “when” (e.g., on Tuesday as opposed to Monday, as well the order of events during the dinner). Thus, impairments in temporal order memory may be related to episodic memory deficits in pre-HD. However, results from the current study demonstrated that impairments in total immediate recall and recognition discriminability in pre-HD near remained statistically significant after controlling for differences in temporal order memory performance. This suggests that deficits on these episodic memory indices in pre-HD could not be explained by impairments on the temporal order memory task. The largest effect sizes in the current study were the group main effects on the temporal order memory task (.43) and on the recognition discriminability index of the HVLT (.45), suggesting that these two measures show the largest group differences.
In addition to impairments on the neuropsychological tasks discussed above, pre-HD near were significantly slower on the visual scanning condition of the TMT. Visual scanning is a considered a fundamental cognitive component of the higher-order executive function condition of the TMT. The present finding of impaired visual scanning in pre-HD is consistent with evidence of impaired volitional occulomotor functioning in pre-HD (Blekher et al., Reference Blekher, Johnson, Marshall, White, Hui and Weaver2006; Reference Blekher, Weaver, Marshall, Hui, Jackson and Stout2008; Golding, Danchaivijitr, Hodgson, Tabrizi, & Kennard, Reference Golding, Danchaivijitr, Hodgson, Tabrizi and Kennard2006). The current study did not detect any other neuropsychological deficits among the pre-HD near, pre-HD far, and normal control groups. These tests included measures of executive functioning, memory, and psychomotor speed. Studies examining neuropsychological function in pre-HD have shown contrasting findings, with some showing impairments in executive functioning, memory, and psychomotor speed in pre-HD (Foroud et al., Reference Foroud, Seimers, Kleindorfer, Bill, Hodes and Norton1995; Kirkwood et al., Reference Kirkwood, Siemers, Hodes, Conneally, Christian and Foroud2000; Lawrence et al., Reference Lawrence, Hodges, Rosser, Kershaw, French-Constant and Rubinsztein1998; Lemiere et al., Reference Lemiere, Decruyenaere, Evers-Kiebooms, Vandenbussche and Dom2002; Paulsen et al., Reference Paulsen, Zhao, Stout, Brinkman, Guttman and Ross2001; Rosenberg et al., Reference Rosenberg, Sorensen and Christensen1995; Wahlin et al., Reference Wahlin, Lundin and Dear2007), but other studies have not found these deficits in pre-HD (Brandt et al., Reference Brandt, Shpritz, Codori, Margolis and Rosenblatt2002; Campodonico, Codori, & Brandt, Reference Campodonico, Codori and Brandt1996). These conflicting findings may be due to several factors such as differences in sample size, prediction equations used to estimate time until disease diagnosis, participants’ estimated years until HD diagnosis, and differential sensitivity of tasks used to assess these cognitive domains. The variable neuropsychological findings in the HD literature point to the need for the development of novel experimental paradigms that may prove to be more sensitive to subtle cognitive changes in pre-HD. The results of the current study suggest that tasks measuring temporal order memory may be sensitive to cognitive changes in pre-HD near diagnosis of HD. Furthermore, manipulations of the temporal separation lag, and therefore temporal interference, on the present task may detect deficits ranging from subtle to severe.
On the present temporal order memory task, deficits in cognitive functions other than temporal order memory may contribute to impaired performance on the task. For example, impairments in visuospatial memory, working memory, or global cognitive function may lead to deficits on the temporal order task. If the deficit in pre-HD near on the present task was due solely to impairments in working memory, visuospatial memory, or global cognitive functioning, then the group would be expected to show a deficit across all temporal separations. However, the present findings showed that pre-HD near matched normal controls and pre-HD far on 6 lag separation trials of the temporal order memory task, in which temporal interference was minimal. Therefore, the impairment in pre-HD near was likely due to a deficit in temporal order memory, and not solely the result of deficits in these other cognitive functions.
In conclusion, the present study found that temporal order memory was impaired in pre-HD close to estimated disease diagnosis relative to pre-HD further from estimated HD diagnosis and normal controls. However, performance improved as a function of decreased temporal interference on the task. The results suggest that the present temporal order memory task may be sensitive to neuropathological dysfunction in pre-HD up to 5 years prior to estimated diagnosis of HD.
ACKNOWLEDGMENTS
We thank Vince Filoteo, Claire Murphy, Scott Roesch, Brianne Bartlett, and Adrienne Callazo for their advice and assistance. We also thank all of the participants in this study for their contributions. The author’s have no conflict of interest. This work was supported by the San Diego State University Grants Program awarded to P.G.
