Memory-guided navigation in amyotrophic lateral sclerosis

Neuropsychological assessment

Seven out of 43 patients (16%) showed performance below ECAS cut-off values in domains relevant for Strong criteria of ALSci [24], i.e. verbal fluency (n = 1), executive function (n = 3) or language (n = 3). However, also in the control group, seven out of 43 subjects showed performance below ECAS cut-off values in executive function (n = 3) or language (n = 4). None of the participants had impairments in more than one domain. Across groups, we found no significant differences for executive function (W = 979.5, p = 0.502), verbal fluency (W = 1003, p = 0.266), language (W = 1019.5, p = 0.131) and visuospatial abilities (W = 927.5, p = 0.761) (descriptive values in Supplementary Table 3). ALS patients showed slightly inferior performance in verbal memory only (W = 1132.5, p = 0.043, effect size = 0.220). Moreover, ALS patients and healthy controls showed comparable performance in visuospatial fluency (5PT number of unique figures: W = 1037.5, p = 0.238), visuospatial memory (SPART immediate and delayed recall: W = 867.5, p = 0.758), and spatial orientation (PTSOT mean angle deviation: W = 882.5, p = 0.861).

Memory-guided navigationHand motor function

Patients and controls did not differ with respect to joystick control abilities as indicated by comparable latency (W = 903, p = 0.857), path length (W = 866, p = 0.618) and velocity (W = 1001, p = 0.514) in the practise task.

Spatial learning in baseline condition

Both groups showed comparable spatial learning abilities as indicated by a similar decrease in latency (repeated-measures ANOVA with Greenhouse–Geisser correction: F(3.814, 316.536) = 0.6356, p = 0.891), path error (repeated-measures ANOVA with Greenhouse–Geisser correction: F(4.091, 339.541) = 0.682, p = 0.574), and search accuracy (repeated-measures ANOVA with Huynh–Feldt correction: F(5.148, 427.284) = 0.803, p = 0.222) across baseline learning trials. Consistent with this observation, the success rate in baseline probe trials did not differ significantly between ALS patients (92.2%) and healthy controls (86.4%) (W = 773.5, p = 0.101). Patients also showed comparable navigation efficiency, as evidenced by similar latency (W = 940, p = 0.898), path error (W = 989, p = 0.582), and search accuracy (W = 926, p = 0.993) in successful baseline probe trials (Fig. 2).

Fig. 2figure 2

Performance in the starmaze task. Blue, ALS group; yellow, control group. First row, initial course of learning across trials in baseline condition. Second to fourth row, navigation performance in successful probe trials. Second row, baseline condition; third row, egocentric condition; fourth row, allocentric condition. Note that both groups show comparable spatial learning abilities and memory-guided spatial navigation performance, as indicated by the lack of group differences in latency, path error, and search accuracy. Data are presented as line plots with standard error and rain cloud plots with individual data points for each participant

Egocentric spatial navigation

Across all egocentric probe trials, 96.5% of ALS patients and 94.0% of healthy controls (W = 898.5, p = 0.937) were able to repeat the correct path sequence and successfully navigated to the correct target location. Navigation efficiency in successful egocentric probe trials was comparable between groups, with similar latency (W = 819, p = 0.714), path errors (W = 793, p = 0.546), and search accuracy (W = 846, p = 0.903) (Fig. 2, Supplementary Fig. 1, Supplementary Table 2). ALS patients thus acquired and used egocentric memory representations as efficiently as healthy controls.

Allocentric spatial navigation

Across all allocentric probe trials, 50.7% of ALS patients and 47.7% of healthy controls applied an allocentric strategy and successfully navigated to the target location (W = 872, p = 0.786). Latency (W = 692, p = 0.796), path error (W = 751, p = 0.732), and search accuracy (W = 770, p = 0.590) in successful allocentric probe trials were also comparable between patients and controls (Fig. 2, Supplementary Fig. 1, Supplementary Table 2). ALS patients thus acquired and used allocentric memory representations as efficiently as healthy controls.

Correlations between memory-guided navigation and clinical variables

We found no correlation of the success rate in the starmaze with disease severity (baseline: rs = − 0.069, p = 0.662, egocentric: rs = − 0.163, p = 0.297, allocentric: rs = − 0.080, p = 0.608) or disease progression (baseline: rs = − 0.146, p = 0.350, egocentric: rs = 0.074, p = 0.639, allocentric: rs = − 0.192, p = 0.218). Similarly, we found no association between navigation parameters of successful trials and disease severity (baseline: latency, rs = − 0.201, p = 0.197; path error, rs = 0.272, p = 0.171; search accuracy, rs = 0.237, p = 0.126; egocentric: latency, rs = − 0.213, p = 0.170; path error, rs = 0.145, p = 0.354; search accuracy, rs = 0.119, p = 0.449; allocentric: latency, rs = − 0.255, p = 0.108; path error, rs = − 0.006, p = 0.970; search accuracy, rs = − 0.001, p = 1.0). For disease progression, we found a correlation only for latency across conditions, but not for other navigation parameters (baseline: latency, rs = 0.544, p < 0.001; path error, rs = 0.091, p = 0.563; search accuracy, rs = 0.033, p = 0.835; egocentric: latency, rs = 0.483, p = 0.001; path error, rs = 0.041, p = 0.792; distance error, rs = 0.010, p = 0.951; allocentric: latency, rs = 0.310, p = 0.049; path error, rs = − 0.109, p = 0.497; search accuracy, rs = 0.016, p = 0.923).

We found no significant differences for any of the investigated navigational variables when ALS patients with normal ECAS values, ALS patients performing below ECAS cut-off values in domains relevant for Strong criteria of ALSci [24], controls with normal ECAS values and controls performing below ECAS cut-off scores were compared (all p ≥ 0.134, see Supplementary Table 4 for details). Furthermore, we found no significant differences for any of the investigated navigational variables when ALS patients, PMA patients, PLS patients and controls were compared (all p ≥ 0.084, see Supplementary Table 5 for details).

Post-navigational memory for maze reconstruction, landmark identity, and landmark positioning

There were no significant differences between ALS patients and healthy controls in reconstruction of the maze layout (W = 924.5, p = 1), recall and identification of environmental landmarks (W = 991, p = 0.568), and positioning of landmarks on the map (W = 1030, p = 0.364).

留言 (0)

沒有登入
gif