The relationship between visuospatial neglect, spatial working memory and search behavior

Figures & data

Table 1. Literature overview. Studies that assessed spatial working memory (SWM) performance in patients with visuospatial neglect (VSN) directly. The total number of participants are split by patients with visuospatial neglect (VSN+), patients without visuospatial neglect (VSN-) and healthy control subjects. In the rightmost column a brief description of the SWM task is provided.

	N
Study	Total	VSN+	VSN-	Control	SWM test
Malhotra et al. (2004)	7	1	3	3	Corsi task; vertical
Pisella et al. (2004)	8	8	-	-	Change detection; 4 × 4 grid with 4 items; delay of 1 s. change in either location, color or shape of 1 item
Malhotra et al. (2005)	20	10	-	10	Corsi task; vertical; identification
Ravizza et al. (2005)	14	1	4	14	N-back task; vertical
Ferber and Danckert (2006)	18	4	-	10	Change detection; vertical; 3 items in right visual field; delay of 3 s; 1 item; same/new location as one of 3 items in sample.
Wansard et al. (2014)	28	14	-	14	Corsi task; increase no. of potential targets to increase difficulty
Wansard et al. (2015)	24	12	-	12	Corsi task; increase no. of potential targets to increase difficulty
Current study	247	24	158	65	Spatial discrimination; vertical; delay of 2 s; distance between first and second stimulus determined with staircase

Table 2. Demographics.

	Control	VSN-		VSN+		Statistics^c
N	65	158 (87%)		24 (13%)		-	-	-
Gender (prop. male)	0.51	0.65		0.46		χ^2(2) = 5.56	-	p = 0.062^d
Age^a	43	59		58		H(2) = 28.4	-	p < 0.001^d
Days post stroke^a	-	21		25.5		U = 14,090	Z = −1.53	p = 0.127
Hemisphere lesioned (prop.)	-		(72.8%)^b		(75.0%)^b	χ^2(2) = 5.03	-	p = 0.081
Left	-	0.50		0.22
Right	-	0.45		0.72
Bilateral	-	0.05		0.06
MoCA^a,e [0–30]	-	23.0	(70.3%)^b	19.0	(75.0%)^b	U = 7642	Z = 2.75	p = 0.004
Motricity Index Arm^a[0–100]	-	87.0	(78.5%)^b	83.0	(79.2%)^b	U = 9145	Z = 1.17	p = 0.183
Motricity Index Leg^a [0–100]	-	91.0	(77.8%)^b	77.0	(79.2%)^b	U = 8948	Z = 0.68	p = 0.340
SAN^a,e [1–7]	-	6	(78.5%)^b	6	(75.0%)^b	U = 8886	Z = 0.25	p = 0.903
Barthel Index^a [0–20]	-	14.0	(79.1%)^b	9.0	(70.8%)^b	U = 9274	Z = 1.56	p = 0.034

^aReported values are medians per group.

^bWe do not have the clinical data of all patients. The percentage of patients for whom we have these data are reported in gray.

^cStatistics are the results of non-parametric tests testing for differences between groups on each parameter.

^dStatistics in this row entails a comparison between all three groups: controls, VSN- and VSN+.

^eMoCA = Montreal Cognitive Assessment; SAN = Stichting Afasie Nederland (Dutch aphasia test).

Figure 1. Trial sequence. Participants had to indicate whether Target 2 (yellow) was presented above or below the location of Target 1 (red). The distance between the locations of Target 1 and Target 2 was controlled by a staircase procedure, i.e. the distance was made smaller when the response on the previous trial was correct, and vice versa. In the delay between Target 1 and Target 2, the fixation point briefly expanded and contracted. Responses were given verbally and entered through a standard keyboard by the experimenter. Actual background color was grey, here depicted as white.

Table 3. Proposed interpretation of Bayes factors.

Bayes Factor (BF10)	Subjective interpretation
>100	Strong evidence in favor of the alternative hypothesis, this result is likely to be replicated
10–100	Substantial evidence in favor of the alternative hypothesis
3–10	Suggestive evidence in favor of the alternative hypothesis, but requires follow-up investigation
⅓–3	Inconclusive, experimental data do not add new information to prior
0.1–⅓	Suggestive evidence against alternative hypothesis, but requires follow-up investigation
0.1–0.01	Substantial evidence against the alternative hypothesis
<0.01	Strong evidence against the alternative hypothesis, this result is likely to be replicated

Figure 2. Spatial working memory (SWM) thresholds. For the boxplots, the central horizontal lines indicate the median threshold per group. Boxes encompass the inter-quartile range, whiskers extend to 2.5^th and 97.5^th quantiles of the group data. Points are participant with scores outside these intervals. (a) Thresholds per group. There was strong evidence in favor of an effect of group on the SWM thresholds (BF₁₀ = 3.12 × 10¹⁴). Moreover, the medians per group are different of the medians of the other groups, as indicated by non-overlapping 95% credible intervals. (b) Thresholds per group and split between males and females. There was no clear evidence for an effect of sex on the SWM thresholds (BF₁₀ = 0.374). (c) Thresholds as a function of age. There was inconclusive evidence for an effect of age on the SWM thresholds (BF₁₀ = 0.934). Ellipses represent area that holds 95% of the cases per group. Blue = control, yellow = VSN-, red = VSN+. Black ellipse represents 95% area across all cases.

Figure 3. Control analyses. Red represents the VSN+ group. Yellow represents the VSN- group. Note that in these analyses we only included a subset of patients for which we had the data (see Table 3). Spatial working memory (SWM) thresholds as a function of (a) MoCA score (BF₁₀ = 141), (b) SAN score, (Dutch aphasia test; BF₁₀ = 0.397), (c) Motricity Index arm (BF₁₀ = 0.490), and (d) Motricity Index leg (BF₁₀ = 0.530), (e) Barthel index (BF₁₀ = 1.16). When accounting for these relations, the VSN+ group still has a higher median SWM threshold than the VSN- group (BF₁₀ = 3.49 × 10⁴). In sum, across patients, a lower MoCA score is related to a higher SWM threshold and the presence of VSN predicts an additional increase in SWM threshold, but none of the other measures for general disability was strongly related to SWM performance.

Table 4. Number of patients with a lateralized attention deficit (VSN) and an SWM impairment (SWM), for whom we also had access to the MoCA score.

	SWM-	SWM+	Total
VSN-	83	15	98
VSN+	3	12	15
Total	86	27	113

Figure 4. Effect of SWM deficit and VSN on search performance, as assessed with the cancellation task with invisible markings. SWM+ group consists of patients with an SWM threshold > 1.24°. Points represents mean, error bars 1 s.e.m. (a) Absolute difference between the number of omission on the left and right side of the search display. There is strong evidence for an effect of VSN on the number of omissions, but inconclusive evidence for an effect of the SWM deficit. (b) Number of delayed revisits. There is substantial evidence for an effect of the SWM deficit on the number of revisits, but inconclusive evidence for a main effect of VSN, or an interaction effect. (c) Best R. There is strong evidence for an effect of SWM, but inconclusive evidence for an effect of VSN or an interaction effect. (d) Intersection rate. There is inconclusive evidence for an effect of either SWM or VSN.

Malhotra, P., Mannan, S., Driver, J., & Husain, M. (2004). Impaired spatial working memory: One component of the visual neglect syndrome. Cortex, 40, 667–676.

 PubMed Web of Science ®Google Scholar

Pisella, L., Berberovic, N., & Mattingley, J. B. (2004). Impaired working memory for location but not for colour or shape in visual neglect: A comparison of parietal and non-parietal lesions. Cortex, 40(2), 379–390.

 PubMed Web of Science ®Google Scholar

Malhotra, P., Jäger, H. R., Parton, A., Greenwood, R., Playford, E. D., Brown, M. M., … Husain, M. (2005). Spatial working memory capacity in unilateral neglect. Brain, 128, 424–435.

 PubMed Web of Science ®Google Scholar

Ravizza, S. M., Behrmann, M., & Fiez, J. A. (2005). Right parietal contributions to verbal working memory: Spatial or executive? Neuropsychologia, 43(14), 2057–2067.

 PubMed Web of Science ®Google Scholar

Ferber, S., & Danckert, J. (2006). Lost in space - The fate of memory representations for non-neglected stimuli. Neuropsychologia, 44(2), 320–325.

 PubMed Web of Science ®Google Scholar

Wansard, M., Meulemans, T., Gillet, S., Segovia, F., Bastin, C., Toba, M. N., & Bartolomeo, P. (2014). Visual neglect: Is there a relationship between impaired spatial working memory and re-cancellation? Experimental Brain Research, 232(10), 3333–3343.

 PubMed Web of Science ®Google Scholar

Wansard, M., Bartolomeo, P., Bastin, C., Segovia, F., Gillet, S., Duret, C., & Meulemans, T. (2015). Support for distinct subcomponents of spatial working memory: A double dissociation between spatial–Simultaneous and spatial–Sequential performance in unilateral neglect. Cognitive Neuropsychology, 32(1), 14–28.

 PubMed Web of Science ®Google Scholar