Brain and Cognition 84 (2014) 69–75

Contents lists available at ScienceDirect

Brain and Cognition journal homepage: www.elsevier.com/locate/b&c

Prose Reading in Neglect Nicoletta Beschin a,b,⇑, Carlo Cisari b, Roberto Cubelli c, Sergio Della Sala d a

Clinical Neuropsychology Unit, Rehabilitation Department, Hospital S. Antonio Abate Gallarate, Varese, Italy Physical and Rehabilitation Medicine, Azienda Universitaria Osped. ‘‘Maggiore della Carità’’, Piemonte Orientale University, Novara, Italy c Department of Psychology and Cognitive Sciences, Università di Trento, Italy d Human Cognitive Neuroscience, Psychology, University of Edinburgh, Edinburgh, UK b

a r t i c l e

i n f o

Article history: Accepted 6 November 2013 Available online 8 December 2013 Keywords: Prose reading Unilateral Spatial Neglect Dyslexia Text neglect Cerebral lesions

a b s t r a c t Prose reading has been shown to be a very sensitive measure of Unilateral Spatial Neglect. However, little is known about the relationship between prose reading and other measures of neglect and its severity, or between prose reading and single word reading. Thirty participants with a first stroke in the right hemisphere and clear symptoms of spatial neglect in everyday life were assessed with tests of prose reading (text in one column book-like, and in two columns magazine-like), single words reading, and a battery of 13 tests investigating neglect. Seventy percent of these participants omitted words at the beginning of the text (left end), showing Prose Reading Neglect (PRN). The participants showing PRN differed from those not showing PRN only for the overall severity of neglect, and had a lesion centred on the insula, putamen and superior temporal gyrus. Double dissociations emerged between PRN and single word reading neglect, suggesting different cognitive requirements between the two tests: parallel processing in single word reading vs. serial analysis in text reading. Notably, the pattern of neglected text varied dramatically across participants presenting with PRN, including dissociations between reading performance of one and two columns text. Prose reading proved a complex and unique task which should be directly investigated to predict the effects of unilateral neglect. The outcome of this study should also inform clinical assessment and advises given to patients and care-givers. Ó 2013 Elsevier Inc. All rights reserved.

1. Introduction Patients with Unilateral Spatial Neglect (USN) (Chica, 2012) may commit errors when reading single words, sentences or texts (Ellis, Flude, & Young, 1987; Riddoch, 1990). This deficit has been termed ‘‘Neglect Dyslexia’’ (ND – for a recent review see Vallar, Burani, & Arduino, 2010) and can be observed also in patients without other manifestations of USN (Haywood & Coltheart, 2000; Warrington, 1990). In word reading, errors include letter deletions (‘‘brain’’ is read ‘‘rain’’), substitutions (‘‘pen’’ is read ‘‘ten’’) and, more rarely, additions (‘‘rose’’ is read ‘‘prose’’), involving the side of the stimulus contralateral to the side of the brain lesion (Arduino, Burani, & Vallar, 2002; Kinsbourne & Warrington, 1962; Lee et al., 2009). Single word reading is affected by perceptual, spatial and lexical factors: the size of the letters and the space between them (Behrmann, Moscovitch, Black, & Mozer, 1990), the stimulus orientation (Nichelli, Venneri, Pentore, & Cubelli, 1993) and its spatial location (Cubelli, Pugliese, & Gabellini, 1994), the lexical status of the ⇑ Corresponding author. Address: Servizio Neuropsicologia, Dipartimento Riabilitazione, Azienda Ospedaliera S. Antonio Abate, Presidio Somma Lombardo, Via Bellini 2, Somma Lombardo, Varese, Italy. E-mail address: [email protected] (N. Beschin). 0278-2626/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.bandc.2013.11.002

stimulus (words tend to be read better than nonwords) and its morphological structure (Behrmann et al., 1990; Hillis & Caramazza, 1991). Research on single word reading has identified different profiles and provided important information to derive theoretical models of word recognition (e.g., Caramazza & Hillis, 1990). ND encompasses also errors committed in reading sentences or texts. However, information about reading complex text is scant and derives mainly from clinical reports or anecdotal observations. In sentence reading, the words at the beginning could be omitted and/or misread (Tegner & Levander, 1991). For instance, the sentence ‘‘In primavera gli uccelli costruiscono il nido’’ [In spring birds make their nest] is read as ‘‘iscono [nonword] il nido’’ (Berti, Ladavas, & Della Corte, 1996) and the sentence ‘‘There is nothing unfair about it’’ is read as ‘‘how about it’’ (Hillis, 2006). The sentence reading task is included in a widely used screening battery for USN (Pizzamiglio, Judica, Razzano, & Zoccolotti, 1989) and appears to be rather sensitive: right brain damaged patients showing left neglect on this task ranged from 39.5% (Massironi, Antonucci, Pizzamiglio, Vitale, & Zoccolotti, 1988) to 41.8% (Zoccolotti et al., 1989). Reading performance is better with meaningful sentences than with meaningless jumble of words (Kartsounis & Warrington, 1989).

70

N. Beschin et al. / Brain and Cognition 84 (2014) 69–75

Paragraph or text reading also has been included in standard diagnostic batteries, e.g. the Behavioural Inattention Test (BIT – Wilson, Cockburn, & Halligan, 1987) or the French Test Battery (Azouvi et al., 2006). Typically, on this task patients do not read more than three (Kartsounis & Warrington, 1989) or five (Làdavas, Paladini, & Cubelli, 1993) words from the extreme right of each line. Text reading has been reported as an extremely sensitive test for left USN (Azouvi et al., 2006; Caplan, 1987; Schwartz, Ojemann, & Dodrill, 1997; Stone et al., 1991): over 46% of right brain damaged patients show left neglect on this task (Azouvi et al., 2006). Different patterns of performance have been reported: some patients omit approximately the same number of words in each line of the text (Kartsounis & Warrington, 1989) whereas other patients tend to produce more omissions and paralexias when reading the last lines than when reading the first ones (Ellis et al., 1987; Vallar, Guariglia, Nico, & Tabossi, 1996). Patients with USN may fail in reading texts but could perform flawlessly in reading single words (Làdavas et al., 1993). Notwithstanding this possible dissociation, it is widely assumed that ND is a unitary disorder; it derives that by assessing single word reading the diagnosis of presence or absence of ND can be made. Indeed, ND is often equated to poor performance on single word processing (e.g., Ward, 2010, p. 268). In particular, little is known about the relationship between prose reading and other measures of USN and its severity, or between prose reading and single word naming. To address these pending issues is the aim of the present study. 2. Materials and methods 2.1. Participants Thirty participants (18 men and 12 women), all Italian speakers, with a first stroke in the right hemisphere were recruited for the study according to the following inclusion criteria: (i) clear symptoms of USN in everyday life (Vossel, Weiss, Eschenbeck, & Fink, 2013) ascertained by means of the Catherine Bergego Scale (CBS), a questionnaire administered to care-givers based on direct observations of the patient’s functioning in 10 daily situations (Azouvi et al., 2002; Azouvi et al., 2006); (ii) availability of morphological neuroimaging to confirm that the brain damage was due to a single lesion and to document its nature and localisation. A total of 18 participants had a haemorrhagic stroke, 12 had an ischemic lesion. The visual field of the participants was assessed clinically by means of the Confrontation Test (Bisiach, Cappa, & Vallar, 1983) and when doubts persisted also by means of a Goldmann Visual Field Test (Siverstone & Hirsch, 1986). Twenty-one of the 30 participants presented with signs of anosognosia for the everyday behavioural effects of their USN ascertained by the difference between their score on the CBS compared to that obtained from the professional caregivers or relatives (Azouvi et al., 2002; Azouvi et al., 2006). None of the patients had any known history of pre-morbid psychiatric or neurological diseases. The participants’ mean age was 63.1 years (sd = 12.7; range: 40–86), they had on average 8.9 years of formal education (sd = 4.2; range = 5–18), and where assessed on average 75 days (sd = 118; range: 15–663) post onset. All participants formally consented to enter the experiment. 2.2. Methods 2.2.1. General neuropsychological examination All participants underwent a brief standardised neuropsychological battery to assess general intellectual and executive abilities by means of verbal tasks. This battery included the Mini-Mental State Examination (Magni, Binetti, Bianchetti, Rozzini, & Trabucchi,

1996), the Verbal Judgement Task (Spinnler & Tognoni, 1987) and the Cognitive Estimation Test (Della Sala, MacPhearson, Phillips, Sacco, & Spinnler, 2003). 2.2.2. Assessment of Unilateral Spatial Neglect Thirteen tests assessing USN were administered to all participants according to the standard procedures. These included: four cancellation tasks, whereby the targets were respectively Stars (Wilson et al., 1987), Lines (Wilson et al., 1987), Letters (Diller, Gerstman, & Gordon, 1974), or Circles (Ota, Fujii, Suzuki, Fukatsu, & Yamadori, 2001); three further cancellation tasks with the stimuli (Lines, Letters or Circles) grouped in two blocks separated by a gap as proposed by Driver and Halligan (1991); four drawing tasks requiring to copy a Complex Scene (Gainotti, Messerli, & Tissot, 1972), a Vase and Two Flowers (Halligan & Marshall, 1993) or Geometrical Shapes (Spinnler & Tognoni, 1987); a Line Bisection test (Wilson et al., 1987); and the verbal Description of a Complex Scene (Cocchini, Cubelli, Della Sala, & Beschin, 1999). The tests were administered in a random order, in one or two sessions depending on the availability of the participant. The cut off score for the diagnosis of USN in each test was taken from the relevant literature or the standardised test manuals. The total number of tests performed below cut-off was used as measure of USN severity; the score of which therefore ranged from zero (no psychometric evidence of USN) to 13 (very severe USN). 2.2.3. Reading tasks 2.2.3.1. Single word reading. Participants were asked to read aloud three blocks of Italian words, each of 35 stimuli (words 5–14 letters long, printed in font Times New Roman size 12). In one block the words were presented one by one in the centre of a single A4 sheet, in the other two blocks, individual words were presented respectively on the right hand side and on the left hand side, along midline of the sheet. Each participant therefore read 35 words in each of three positions. The three blocks were presented randomly. According to the criteria of Ellis et al. (1987), paralexias were classified as left neglect errors if the target and the response were identical to the right of an identifiable neglect point and there were no letters in common to the left of this point. These errors denote Single Word Neglect (SWN). 2.2.3.2. Prose reading. Passages of prose were presented in two formats, book-style, i.e. with the text running continuously on one column, and magazine style, i.e. with the text distributed in two columns separated by a gap. The text was in Times New Roman 12. 2.2.4. Text on one column Participants were asked to read aloud a short prose passage comprising 130 words distributed in 11 lines, each 142 mm long. Caplan (1987) posited that reading with both the left and right margin irregularly indented, i.e. less predictable, might prove more sensitive to the effect of USN (see also Bachman, Fein, Davenport, & Price, 1993). However, this assumption did not hold when experimentally tested (Towle & Lincoln, 1991). Moreover, we wanted the text to be as close as possible to a real text as encountered in everyday life. Therefore, for the purpose of this study, we used a standard, fully justified format. Following Caplan (1987), participants were diagnosed as having Prose Reading Neglect (PRN) when they omitted sequences of words or individual words at the beginning (on the left) of at least one of the rows. Even the omission of the first letter of the first word was considered as indicative of PRN, provided that was not the only error type occurring, otherwise it would have been difficult to disentangle it from the effect of SWN. The length of omission in mm, and the number of incomplete lines were taken as measures of severity. These two measures captured the two spatial

71

N. Beschin et al. / Brain and Cognition 84 (2014) 69–75

dimensions of possible reading deficits, horizontal and vertical, respectively. Omission in mm was preferred as a measure over number of words omitted, as words can be of different length, hence possibly biasing the comparison across lines and the gradient analysis. Within the attended text, errors on individual words were considered according to the same criteria detailed above (see Single Word Reading). A word omitted between two read words was also classed as individual word error rather than evidence of PRN. 2.2.5. Text on two columns Participants were asked to read aloud a short prose passage on 2 columns, comprising 130 words (62 in the left column and 68 in the right column) distributed in 11 lines. In the left column every row was 78 mm long except the last one which was 45 mm. In the right column each row was 78 mm long, except the last one which was 70 mm. Participants were diagnosed as having PRN when they omitted words or letters at the beginning (on the left) of at least one of the rows on either of the two columns, or omit a whole column. The same scoring and diagnostic criteria as above apply. 2.3. Lesion analysis All patients’ lesions as detected by morphological neuroimaging (CT or MRI scans) have been mapped one by one using the freely available programme MRIcro (Rorden & Brett, 2001) which also allows a direct subtraction of lesion overlap of one group of participants from another group. The scan mapped was the closest to the assessment, always preceded the assessment and the interval between the dates of scan and assessment never exceeded 3 weeks. 3. Results 3.1. Text on one column A total of 21/30 participants (70%) omitted words or letters at the beginning of the text (left end), hence showing PRN (PRN+), whereas 9 patients (30%) read the prose passage without missing the beginning of any lines (PRN ). The demographic features and neuropsychological performance of PRN+ and PRN participants are detailed in Table 1, together with information about presence-absence of any field defect information about their lesion. The two subgroups did not differ significantly in age, education, days post-onset, or MMSE, Verbal Judgments and Cognitive Estimations scores (all ts < 1). The average USN severity score for the entire group of 30 participants was 4.5 (range: 0–13, median = 3). PRN+ participants presented with USN (mean = 9.11 (sd 2.84), range: 1–13, median = 12) which was more severe (t = 1.90, p < 0.05) than PRN participants (mean = 2.52 (3.35), range: 0–12, median = 5) (see Table 1). It is noteworthy that all PRN+ participants performed poorly at least on one of the 13 formal tests used to assess USN. The Pearson correlation between prose reading (mm of text omitted)

and neglect severity was significant (r = 0.52, p < .05). By scrutinising the individual performance in the USN tests it appears that PRN+ participants showed often signs of USN in the Star Cancellation (20/21), and in the Letter Cancellation in two blocks (20/21), whereas the tests least associated with PRN were the Line Cancellation, and Line Bisection, failed only by 12 and 13 PRN+ participants respectively. However, this outcome reflects the test sensitivity to USN, as the Star Cancellation and the Letter Cancellation in two blocks were also failed by 6/9 PRN whereas the Line Cancellation and Line Bisection only by 2/9. In sum, no individual test assessing USN could reliably predict the presence of PRN. Anosognosia for USN was present in 16/21 PRN+ (76%) and 5/9 PRN (55%) participants, suggesting no association between unawareness of USN and presence of PRN. PRN+ participants on average attended to 108.36 mm (range: 3.09–141.8 mm) of text, omitting text on the left end side on average in 5.47 lines (range: 1–11). On the whole, these participants read in a similar way the upper part of the text (first five lines, mean = 109.77, sd = 44.25) and the lower part (last five lines, mean = 106.21, sd = 49.68). However, different individuals presented with different patterns of prose reading (see Fig. 1). Six participants had a particular severe PRN as they omitted the initial words or letters in all the 11 lines. They did show a top–bottom gradient, with the upper part of the text (first five lines, mean = 63.14, sd = 59.09) read more accurately than the lower part (last five lines, mean = 49.73, sd = 55.01). This difference is reliable (p < .05). As reported in Fig. 2, there was a significant correlation between the number of lines read entirely and the number of passed tests for visuo-spatial neglect (r = .704, p < .05). Eight participants (27%) read all the words in the Single Word Reading test without showing SWN, the other 22 presented with a range of errors from 1 to 96 out of 105 words. Seventeen patients (57%) showed both SWN and PRN. However, these two reading impairments double dissociated. There were 5 patients (17%) with SWN but no PRN whereas 4 (13%) had the opposite pattern presenting with PRN but not SWN. Only four participants had neither SWN nor PRN. In reading single words, participants with PRN+ committed on average more errors than PRN participants (Table 2). In prose reading, thirteen participants (43%) did not show neglect errors in reading individual words in prose reading, but seven of them showed prose neglect reading. As shown by Table 2, PRN+ and PRN did not differ in producing neglect errors (t = 0.16, n.s.) and non-neglect errors (t = 0.32; n.s.). As a partial caveat to these observations, it is worth noting that the total number of words read by PRN+ participants is less than those read by the PRN participants. Paralexias often involved words on both sides of the text (see Fig. 3). Indeed, the average number of reading errors in the left sided 71 mm of text did not differ from that observed in the right sided 71 mm, PRN+ participants committed 2.23 errors (sd = 2.71) on the left side compared to 2.23 (sd = 2.42) on the right, whereas PRN patients committed 2.55 (sd = 4.58) on the left side and 1.66 (sd = 2.12) on the right.

Table 1 Demographic features and neuropsychological performances of the participants subdivided in those showing (PRN+) and those not showing (PRN ) Prose Reading Neglect; Standard deviations in parentheses.

PRN+ (n = 21) PRN (n = 9)

Age

Education (years)

Sex (M/ F)

Days postonset

Type of lesion

Visual field

MMSE (0– 30)

Verbal Judgement (0–60)

Estimation test (0–80)

USN severity (0– 13)

62.19 (12.85) 65.11 (12.76)

8.90 (4.23) 9.00 (4.35)

11/10

82.10 (138.68) 55.44 (24.99)

15Hm 6I 3Hm 6I

4H;2Q 10E;5N 2H; 4E;3N

23.29 (4.34) 24.53 (4.06)

38.62 (10.95) 45,00 (12.39)

17.90 (4.40) 18,11 (3.29)

10.48 (3.35) 3.89 (2.84)

6/3

M = Male; F = Female; Hm = Haemorrage; I = Ischaemia; H = Haemianopsia; Q = Quadrantopsia; E = Extinction; N = Normal.

72

N. Beschin et al. / Brain and Cognition 84 (2014) 69–75

Case 20

Lines

Lines

Case 12 1 2 3 4 5 6 7 8 9 10 11 0

50

100

1 2 3 4 5 6 7 8 9 10 11 0

Line length in mm

20

40

60

80

100

120

140

Line length in mm

Fig. 1. Two examples of participants showing contrasting reading patterns. Solid black lines represent omitted text, white lines text read independently of accuracy. Both participants presented with USN in all the 13 tests of the diagnostic battery (see Section 2.2).

Fig. 2. Number of USN tests performed above cut-off scores (i.e., normally) and number of text lines entirely read by each individual participant identified by their protocol number.

Table 2 Errors produced by the participants showing (PRN+) and not showing (PRN ) Prose Reading Neglect in single word reading task and prose reading task (standard deviations in parentheses). PRN+ (n = 21)

PRN

Single word reading Total number Neglect errors Non neglect errors

(n = 9)

36.0 (39.7) 29.7 (37.6) 6.3 (3.0)

4.7 (5.4) 2.2 (4.9) 2.4 (2.5)

Prose reading Total number Neglect errors Non neglect errors

4.95 (3.6) 2.8 (3.6) 2.1 (3.1)

4.8 (6.7) 1.0 (1.7) 3.8 (5.8)

PRN subgroup, although no lesion resulted specific for either group. Within the PRN+ subgroup 86% of the bins share a core localisation centred upon the insula, putamen, superior temporal lobe and Rolandic operculum. Within the PRN subgroup a common lesion encroaching upon the insula is covered by 86% of the bins whereas 80% of the bins include both the Rolandic operculum and the insula. Therefore, there was no notable anatomical difference in the lesions sites between the two groups. This was confirmed by the comparison intersection between the two groups which failed to show between groups differences.

3.2. Text on two columns

Fig. 3. An example from a participant, Case 2, showing paralexias involving words in both spatial sides. Underlined the instances of prose reading neglect (omitted portions of individual lines), in bold the paralexias (single words misread within the attended text).

The outcome of the analysis of the lesions is shown in Fig. 4. The mask analysis carried out with MRIcro (Rorden & Brett, 2001) showed that the PRN+ had a lesion more extensive than the

A total of 19/30 participants (63.3%) showed PRN on the two columns test. Twelve participants showed PRN in both columns, five only in the left sided column, and two only in the right sided column (Cases 2 and 15), though their PRN was minimal and centred solely on lines 2 and 3, and on lines 3 and 4, respectively. Of the 21 showing PRN with one column text, 17 also presented with PRN with two columns; 2 participants had PRN with two columns but not with the one column version of the test. Notably, while Case 15 performed flawlessly on the one column text, Case 2 (see Fig. 4) showed a clear PRN also on the one column text. Participants showing PRN on two columns, on average attended to 48.64 mm (sd = 33.39, range: 0–75 mm) and 60.99 mm (sd = 26.48, range: 5.63–77.27) of text of the left and the right column, respectively (p = .02). They omitted text on the left end side of the left column on average in 5.36 lines (sd = 4.99, range: 0–11), and omitted text on the left end side of the right column on average in 3.42 lines (sd = 4.70, range: 0–11). One participant (Case 30) neglected the entire left column while performing flawlessly on the right column, and performed the one

N. Beschin et al. / Brain and Cognition 84 (2014) 69–75

Fig. 4. In the upper panel the overlap lesion plots of the 21 PRN+ patients is shown. In the central panel the overlap lesion plots of the 9 PRN patients is shown. Overlap map show the degree of involvement for each voxel in the lesions, normalised to the template. Each individual lesion was superimposed onto a 2D axial rendering on the MNI representative brain in stereotactic space. The numbered slices correspond to Z-coordinates. Eight axial slices are shown corresponding to the Z-coordinates 40, 32, 24, 16, 8, 0, 8, 16. The range of the colour scale derives from the absolute number of patient lesions. On the overlap maps, the regional frequency of brain lesions in each area of the right hemisphere is expressed according to the colour scale from violet to red (from 1 to 21 and from 1 to 9 respectively for PRN+ and PRN ). The third panel shows the contrast map of the relative involvement of each voxel in the lesions of PRN+ group minus the PRN group. Colour scale indicates specificity of lesion localisation, whereby violet (point zero) represents complete lack of specificity and the scale ranges from involvement only in the PRN subgroup (light blue) to involvement only in the PRN+ subgroup (light red). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

column text errorless. This contrasts with the performance of Case 20 who showed a severe PRN on the one column text (see Fig. 1) coupled with showing severe PRN on the left column in the two columns test. Within the egocentric co-ordinates the dichotomy has been proposed between space and object based neglect (Marshall & Halligan, 1993; Walker, 1995). Accordingly, case 12 shows a reading pattern indicative of space-based neglect, as he ignores the entire left column, shows PRN on the left side of the right column and the left side of the one column text, in both cases with an increasing top-down gradient. On the contrary other participants showed a pattern of PRN virtually overlapping on the two boxes, neglecting left sided text both on the left and the right column which is indicative of a clear object based USN, for instance Case 27 who also neglected the left side of the one column text. 4. Discussion Prose reading is sometimes described anecdotally or in single case reports (e.g. Kartsounis & Warrington, 1989), is recommended as a sensitive test of USN (Azouvi et al., 2006; Caplan, 1987; Schwartz et al., 1997; Stone et al., 1991) and is included in some test batteries for the diagnosis of Unilateral Spatial Neglect (USN) (e.g., Azouvi et al., 2006; Halligan, Marshall, & Wade, 1989). However, prose reading in patients with USN has never been sys-

73

tematically investigated. Here we showed that two thirds of the patients with USN presented with Prose Reading Neglect (PRN), that is, they omitted the beginning of lines in reading passages of text. The pattern of neglected text presented by patients with USN is clearly different from that presented by patients with Hemianopic Dyslexia, who typically show few errors centred on individual words within the text body rather than omitting chunks of text from the left side and show meaningful completions of words which, at odds with Neglect Dyslexia, could be either at their beginning or ending (see review in Schuett, Heywood, Kentridge, & Zihl, 2008). The anatomical substrates of PRN are centred on the insula, putamen, superior temporal lobe and Rolandic operculum. These locations are mostly overlapping with those reported by Karnath and colleagues as typical core lesions associated with USN (Karnath, Ferber, & Himmelbach, 2001; Karnath & Himmelbach, 2002; Karnath, Himmelbach, & Rorden, 2002; see also e.g., Chechlacz et al., 2013; but see for a different view e.g., Marshall, Fink, Halligan, & Vallar, 2002). However, no specific lesion emerged which differentiates PRN+ from PRN . Similarly, no specific tests assessing USN could predict the presence of PRN. Clear double dissociations emerged between PRN and SWN indicating that a full profile of each patient’s performance is needed to characterise the pattern of their spared and impaired reading abilities. This dissociation could be accounted for in terms of the different visual processing requirements between the two tests: parallel in single word reading vs. serial in text reading. This explanation resonates that offered by Cocchini et al. (1999) to account for the profile shown by their patient AB and others reported in the literature (e.g., Kartsounis & Findley, 1994), who presented with clear signs of USN in visual search and cancellation tasks but not in tasks like line bisection or the Wund-Jastrow Illusion test. Reading single words implies the parallel processing of letter groupings to be treated and recognised as unitary objects, which can be fully detected within a single fixation; on the contrary, reading prose requires gaze movements for spatial exploration and searching strategies for meaning and coherence. Consequently, semantic inferences are effectual in prose reading though irrelevant in single word reading. Hence, a participant with word recognition disorders (Caramazza & Hillis, 1990) might show SWN yet perform well on prose reading thanks to contextual cues (both semantic and syntactic). On the other hand a participant with deficits in space exploration (Bisiach, Geminiani, Berti, & Rusconi, 1990) might show PRN but perform well in Single Word Reading. The dissociating performances with the two variants of the reading tests, with one and two columns, suggest that in clinical practice it would be advisable to assess both formats to diagnose the presence of PRN. This can have serious implications in counselling patients with USN and their care-givers, as it may be relevant in adjusting the patients’ reading ability in their daily activities, for instance by selecting and adapting the different format of reading material to their optimal performance, adding to their comprehension and enjoyment. In the current study, a significant correlation between severity of PRN (measured as the number of lines with the beginning omitted) and severity of USN (measured as the number of tests for spatial exploration and processing performed below cut-off) was found. Patients omitting the left side of all the lines of the prose reading test appeared to fail almost all clinical tests for USN. Therefore, the number of lines entirely read can be assumed as an index of severity of visuo-spatial neglect: the more lines are misread in the left end side, the more severe is the clinical manifestation of neglect syndrome. Overall participants show no top–bottom gradient in reading individual lines, however patterns of individual differences clearly surfaced; when only the most severe cases were considered, a clear

74

N. Beschin et al. / Brain and Cognition 84 (2014) 69–75

gradient pattern emerged, showing overwhelming neglect of the left side of the bottom lines, supporting the view that severe USN might not be equally distributed in the left hemispace, rather affecting mostly the left inferior quadrant (Halligan & Marshall, 1989; Morris, Mickel, Brooks, Swavely, & Heilman, 1985). Alternatively, a patient who makes errors might increasingly lose their grasp on the meaning of the text, so when reading lower lines, the text might lose coherence and they might have less motivation to try making sense of it. This finding suggesting a possible topdown gradient, may also have relevance in clinical settings, as it is custom to curtail prose reading tests half way through when the first half is read flawlessly or when ‘‘many errors’’ are observed (Lezak, 2004, p. 383). If PRN can be considered evidence of a severe USN, reading texts flawlessly does not necessarily entail absence of other forms of USN. In our sample two patients showed dissociated behaviour: Case 5 read correctly the whole text but performed normally only four (out of thirteen) clinical tests for USN; Case 11 omitted the beginning of only one line in prose reading but showed USN in eight tests. It is worth noting that both patients were fully aware of the presence of their neglect disorder. It might be possible that awareness of the disease induced patients to overcome their deficits by taking advantage from all cues available in texts, that are both verbal (meaning and coherence) and spatial (straight margin on the left). In sum, based on this pivotal study, prose reading can no longer be thought of as a marginal task whose performance could be inferred by other measures. On the contrary, it should be considered within assessing test batteries, as PRN carries specific information useful both for laboratory studies and clinical practice, adding to our understanding and managing of the behavioural difficulties shown by patients affected by USN. Acknowledgment We are grateful to Rob McIntosh who critically read an earlier version of this manuscript. References Arduino, L. S., Burani, C., & Vallar, G. (2002). Lexical effects in left neglect dyslexia: A study in Italian patients. Cognitive Neuropsychology, 19(5), 421–444. Azouvi, P., Bartolomeo, P., Beis, J., Perennou, D., Pradat-Diehl, P., & Rousseaux, M. (2006). A battery of tests for the quantitative assessment of unilateral neglect. Restorative Neurology and Neuroscience, 24, 273–285. Azouvi, P., Samuel, C., Louis-Dreyfus, A., Bernati, T., Bartolomeo, P., Beis, J. M., et al. (2002). Sensitivity of clinical and behavioural tests of spatial neglect after right hemisphere stroke. Journal of Neurology, Neurosurgery, and Psychiatry, 73, 160–166. Bachman, L., Fein, G., Davenport, L., & Price, L. (1993). The indented paragraph reading test in the assessment of left hemi-neglect. Archives Clinical Neuropsychology, 8(6), 485–496. Behrmann, M., Moscovitch, M., Black, S. E., & Mozer, M. (1990). Perceptual and conceptual mechanisms in neglect dyslexia. Brain, 113, 1163–1183. Berti, A., Ladavas, E., & Della Corte, M. (1996). Anosognosia for hemiplegia, neglect dyslexia, and drawing neglect: Clinical findings and theoretical considerations. Journal of the International Neuropsychological Society, 2, 426–440. Bisiach, E., Cappa, S., & Vallar, G. (1983). Guida all’esame neuropsicologico. Milano: Cortina. Bisiach, E., Geminiani, G., Berti, A., & Rusconi, M. L. (1990). Perceptual and premotor factors of unilateral neglect. Neurology, 40, 1270–1281. Caplan, B. (1987). Assessment of unilateral neglect: a new reading test. Journal of Clinical and Experimental Neuropsychology, 9, 359–364. Caramazza, A., & Hillis, A. E. (1990). Levels of representation, co-ordinate frames, and unilateral neglect. Cognitive Neuropsychology, 7, 391–445. Chechlacz, M., Rotshtein, P., Hansen, P. C., Deb, S., Riddoch, M. J., & Humphreys, G. W. (2013). The central role of the temporo-parietal junction and the superior longitudinal fasciculus in supporting multi-item competition: Evidence from lesion-symptom mapping of extinction. Cortex, 49, 487–506. Chica, A. B., Thiebaut de Schotten, M., Toba, M., Malhotra, P., Lupiáñez, J., & Bartolomeo, P. (2012). Attention networks and their interactions after righthemisphere damage. Cortex, 48, 654–663. Cocchini, G., Cubelli, R., Della Sala, S., & Beschin, N. (1999). Neglect without extinction. Cortex, 35, 285–313.

Cubelli, R., Pugliese, M., & Gabellini, A. S. (1994). The effect of space location on neglect depends on the nature of the task. Journal of Neurology, 241, 611–614. Della Sala, S., MacPhearson, S. E., Phillips, L. H., Sacco, L., & Spinnler, H. (2003). How many camels are there in Italy? Cognitive estimates standardised on the Italian population. Neurological Sciences, 24, 10–15. Diller, L., BenYishay, Y., Gerstman, L. J., Goodkin, R., Gordon, W., & Weinberg, M. S. (1974). Studies in cognition and rehabilitation in Hemiplegia. New York: University Medical Center. Rehabilitation Monograph n.50. Driver, J., & Halligan, P. W. (1991). Can visual neglect operate in object-centered coordinates? An affirmative single-case study. Cognitive Neuropsychology, 8, 475–496. Ellis, A. W., Flude, B. M., & Young, A. W. (1987). ‘‘Neglect dyslexia’’ and the early visual processing of letters in words and non words. Cognitive Neuropsychology, 4, 439–464. Gainotti, G., Messerli, P., & Tissot, R. (1972). Qualitative analysis of unilateral spatial neglect in relation to laterality of cerebral lesions. Journal of Neurology Neurosurgery and Psychiatry, 35, 545–550. Halligan, P. W., & Marshall, J. C. (1989). Is neglect (only) lateral? A quadrant analysis of line cancellation. Journal of Clinical and Experimental Neuropsychology, 11, 793–798. Halligan, P. W., & Marshall, J. C. (1993). When two is one: a case study of spatial parsing in visual neglect. Perception, 22, 309–312. Halligan, P. W., Marshall, J. C., & Wade, D. T. (1989). Visuospatial neglect: Underlying factors and test sensitivity. The Lancet, 2, 908–911. Haywood, M., & Coltheart, M. (2000). Neglect dyslexia and the early stages of visual word recognition. Neurocase, 6, 33–44. Hillis, A. E. (2006). Neurobiology of unilateral spatial neglect. Neuroscientist, 12(2), 153–163. Hillis, A. E., & Caramazza, A. (1991). Spatially-specific deficit to stimulus-centered letter shape representations in a case of ‘‘unilateral neglect’’. Neuropsychologia, 29, 1223–1240. Karnath, H. O., Ferber, S., & Himmelbach, M. (2001). Spatial awareness is a function of the temporal not the posterior parietal lobe. Nature, 411, 950–953. Karnath, H. O., & Himmelbach, M. (2002). Strategies of lesion localization. Reply to Marshall, Fink, Halligan and Vallar. Cortex, 38, 258–260. Karnath, H. O., Himmelbach, M., & Rorden, C. (2002). The subcortical anatomy of human spatial neglect: putamen, caudate nucleus, and pulvinar. Brain, 125, 350–360. Kartsounis, L. D., & Findley, L. J. (1994). Task specific visuospatial neglect related to density and salience of stimuli. Cortex, 30, 647–659. Kartsounis, L. D., & Warrington, E. K. (1989). Unilateral visual neglect overcome by cues implicit in stimulus arrays. Journal of Neurology, Neurosurgery and Psychiatry, 52, 1253–1259. Kinsbourne, M., & Warrington, E. K. (1962). A variety of reading disability associated with right-hemisphere lesion. Journal of Neurology, Neurosurgery and Psychiatry, 25, 339–344. Làdavas, E., Paladini, R., & Cubelli, R. (1993). Implicit associative priming in a patient with left visual neglect. Neuropsychologia, 31, 1307–1320. Lee, B. H., Suh, M. K., Kim, E., Seo, S. W., Choi, K. M., Kim, G., et al. (2009). Neglect dyslexia: Frequency, association with other hemispatial neglects and lesion localization. Neuropsychologia, 47(3), 704–710. Lezak, M. D. (2004). Neuropsychological assessment (4th ed.). New York: Oxford University Press. Magni, E., Binetti, G., Bianchetti, A., Rozzini, R., & Trabucchi, M. (1996). Mini-mental state examination: A normative study in Italian elderly population. European Journal of Neurology, 3, 198–202. Marshall, J. C., Fink, G. R., Halligan, P. W., & Vallar, G. (2002). Spatial awareness: A function of the posterior parietal lobe? Cortex, 38, 253–257. Marshall, J. C., & Halligan, P. W. (1993). Visuo-spatial neglect: A new copying test to assess perceptual parsing. Journal of Neurology, 240(1), 37–40. Massironi, M., Antonucci, G., Pizzamiglio, L., Vitale, M. V., & Zoccolotti, P. (1988). The Wundt-Jastrow illusion in the study of spatial hemi-inattention. Neuropsychologia, 26, 161–166. Morris, R., Mickel, S., Brooks, M., Swavely, S., & Heilman, K. M. (1985). Recovery from neglect. Journal of Clinical and Experimental Neuropsychology, 7, 609. Nichelli, P., Venneri, A., Pentore, R., & Cubelli, R. (1993). Horizontal and vertical neglect dyslexia. Brain and Language, 44, 264–283. Ota, H., Fujii, T., Suzuki, K., Fukatsu, R., & Yamadori, A. (2001). Dissociation of bodycentered and stimulus-centered representations in unilateral neglect. Neurology, 57, 2064–2069. Pizzamiglio, L., Judica, A., Razzano, C., & Zoccolotti, P. (1989). Toward a comprehensive diagnosis of visual-spatial disorders in unilateral brain damaged patients. Evaluacion Psicologica/Psychological Assessment, 5, 313–320. Riddoch, M. J. (1990). Neglect and the pheripheral dyslexias. Cognitive Neuropsychology, 7, 369–389. Rorden, C., & Brett, M. (2001). Stereotaxic display of brain lesions. Behavioural Neurology, 12, 191–200. Schuett, S., Heywood, C. A., Kentridge, R. W., & Zihl, J. (2008). The significance of visual information processing in reading: Insights from hemianopic dyslexia. Neuropsychologia, 46, 2445–2462. Schwartz, T. H., Ojemann, G. A., & Dodrill, C. B. (1997). Reading errors following right hemisphere injection of sodium amobarbital. Brain Language, 58, 70–91. Siverstone, D. E., & Hirsch, J. (1986). Automated visual field testing. Norwalk, CT: Appelton-Century Croft. Spinnler, H., & Tognoni, G. (1987). Standardizzazione e taratura italiana di test neuropsicologici. The Italian Journal of Neurological Sciences (Suppl. 8), 1–120.

N. Beschin et al. / Brain and Cognition 84 (2014) 69–75 Stone, S. P., Wilson, B., Wroot, A., Halligan, P. W., Lange, L. S., Marshall, J. C., et al. (1991). The assessment of visuo-spatial neglect after acute stroke. Journal Neurology Neurosurgery and Psychiatry, 54, 345–350. Tegner, R., & Levander, M. (1991). Through a looking glass. A new technique to demonstrate directional hypokinesia in unilateral neglect. Brain, 114, 1943–1951. Towle, D., & Lincoln, N. B. (1991). Use of the indented paragraph test with right hemisphere-damaged stroke patients. British Journal Clinical Psychology, 30, 37–45. Vallar, G., Burani, C., & Arduino, L. S. (2010). Neglect dyslexia: A review of the neuropsychological literature. Experimental Brain Research, 206, 219–235. Vallar, G., Guariglia, C., Nico, D., & Tabossi, P. (1996). Left neglect dyslexia and the processing of neglected information. Journal Clinical Experimental Neuropsychology, 18, 733–746.

75

Vossel, S., Weiss, P. H., Eschenbeck, P., & Fink, G. R. (2013). Anosognosia, neglect, extinction and lesion site predict impairment of daily living after righthemispheric stroke. Cortex, 49, 1782–1789. Walker, R. (1995). Spatial and object base neglect. Neurocase, 1, 371–383. Ward, J. (2010). The student’s guide to cognitive neuroscience (2nd ed.). Hove: Psychology Press. Warrington, E. K. (1990). Right neglect dyslexia: A single case study. Cognitive Neuropsychology, 8, 193–212. Wilson, B., Cockburn, J., & Halligan, P. (1987). Behavioral Inattention Test. Tichfield, Hants: Thames Valley Test Company. Zoccolotti, P., Antonucci, G., Judica, A., Montenero, P., Pizzamiglio, L., & Razzano, C. (1989). Incidence and evaluation of the hemi-neglect disorder in chronic patients with unilateral right brain damage. International Journal of Neuroscience, 47, 209–216.