- Email: [email protected]
Visual deficits in pre-readers at familial risk for dyslexia
Vision Research 48 (2008) 2835–2839
Contents lists available at ScienceDirect
Vision Research journal homepage: www.elsevier.com/locate/visres
Visual deficits in pre-readers at familial risk for dyslexia Alison Kevan *, Kristen Pammer The Australian National University, School of Psychology, Canberra ACT 0200, Australia
a r t i c l e
i n f o
Article history: Received 6 August 2008 Received in revised form 26 September 2008
Keywords: Dorsal stream Magnocellular Dyslexia Visual frequency doubling Coherent motion Pre-reading children
a b s t r a c t Visual processing deficits in dyslexic readers are argued to evolve as a consequence of reading failure. This study examines dorsal stream functioning of children before they commence formal reading instruction to determine whether visual deficits precede reading difficulties. Coherent motion and visual frequency doubling detection were measured in children at familial risk for dyslexia and in children unselected for family reading history. Here we show that children who are at family risk for dyslexia demonstrate dorsal stream deficits before they learn to read, whilst demonstrating no corresponding deficits in coherent form and static grating control tasks. Results indicate that the dorsal visual deficits observed in dyslexic readers are unlikely to be the result of reading failure. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction The most widely acknowledged impairment in dyslexia is a difficulty in phonological processing. As such dyslexia is fundamentally considered to be a language based problem in the cognitive manipulation of words (Vellutino, Fletcher, Snowling, & Scanlon, 2004). However, it is now clear that a phonological account on its own cannot explain the full extent of impairments observed in dyslexic readers. Thirty years of research indicates that many dyslexic readers also possess a visual processing deficit that is specific to the magnocellular dominated dorsal pathway. Convergent evidence from behavioral (e.g., Cornelissen, Richardson, Mason, Fowler, & Stein, 1995; Kevan & Pammer, 2008; Pammer & Kevan, 2007; Pammer & Vidyasagar, 2005; Pammer & Wheatley, 2001; Talcott, Hansen, Assoku, & Stein, 2000; Wilmer, Richardson, Chen, & Stein, 2004), anatomical (e.g., Galaburda & Livingstone, 1993; Livingstone, Rosen, Drislane, & Galaburda, 1991) and imaging studies (e.g., Demb, Boynton, Best, & Heeger, 1998a; Demb, Boynton, & Heeger, 1998b; Eden et al., 1996) provide support to suggest that dyslexic readers are less sensitive to visual stimuli that is mediated by the dorsal stream. These findings have led to theoretical models that explain the link between a dorsal stream deficit and reading, including attention (Vidyasagar, 2004), ocular motor control (Stein, 2001), and letter position encoding (Whitney & Cornelissen, 2005), see Boden and Giaschi (2007) for a recent review. The suggestion of a dorsal stream deficit in dyslexic readers that is related to reading failure however is controversial (Ramus, 2001,
* Corresponding author. Fax: +61 2 6125 0499. E-mail address: [email protected] (A. Kevan). 0042-6989/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.visres.2008.09.022
2003; Skottun, 2000). In fact, it is counter-intuitive that the dorsal (visual ‘‘where”) stream should have a role in reading as the mechanisms for the detailed feature analysis required for reading should lie with the ventral (visual ‘‘what”) stream, which has a higher spatial resolution and a role in object recognition. Thus visual deficits are frequently dismissed as being secondary to a core language deficit, where dorsal stream impairments develop as a consequence of failing to learn to read (Ramus, 2003; Vellutino et al., 2004), as the reading network does not receive adequate visual exposure from print and thus may not provide sufficient input to the system to stimulate neural connections. Indeed, it is biologically plausible that such reciprocal feedback could occur. The reading network is highly interactive and dynamic, requiring the synthesis of signals from disparate areas of the brain, and learning to read requires integrating and tuning different components of that network in a highly sophisticated way. Thus, Hebbian learning principles would predict that impairment in one part of the network (e.g. a core language deficit) could result in failure to develop other components of the network, (e.g., visuo-spatial encoding). Therefore it is currently unknown whether the dorsal stream deficits are a cause or a consequence of reading failure. However, if it could be demonstrated that dorsal stream deficits exist in children at family risk of dyslexia prior to the onset of reading, this would challenge the argument that dorsal stream deficits are the result of reading failure. Numerous studies have demonstrated that dyslexic readers are less sensitive to coherent motion than control readers (Cornelissen et al., 1995; Hulslander et al., 2004; Pammer & Wheatley, 2001; Talcott et al., 2000; Wilmer et al., 2004). This is in contrast to normal performance on static tasks measuring coherent form detection (Hulslander et al., 2004; Wilmer et al., 2004). This dissociation suggests that dyslexic readers possess an impairment specific to the
2836
A. Kevan, K. Pammer / Vision Research 48 (2008) 2835–2839
Magno Pathway
Parvo Pathway
Retina
LGN
M cells
Magno
Layer 4Cα
Parvo
Layer 4C β
P cells
V1
V2 Layer 4β
Thick Stripes
Blob
Thin Stripes
InterBlobs
Extra striate
InterStripes
V5/MT
Dorsal Stream
V4
Ventral Stream
Fig. 1. Representation of the parallel pathways in the primate visual system. Lines indicate connections between components. From V1 magno and parvocellular information interact considerably as they project to extrastriate visual areas (Merigan & Maunsell, 1993; Vidyasagar et al., 2002), suggesting that higher order dorsal processing may not be entirely indicative of lower level magno functioning. LGN: lateral geniculate nucleus, MT: middle temporal area. Adapted from Merigan and Maunsell (1993).
detection of the dynamic properties of visual stimuli, and adds further support to the notion of a dorsal stream deficit. However, support for the use of coherent motion is not universal (Skottun & Skoyles, 2006), as equivocal results have been demonstrated in dyslexic children (Kronbichler, Hutzler, & Wimmer, 2002) and in prereaders1 (Boets, Wounters, van Wieringen, & Ghesquière, 2006). Given that magnocellular and parvocellular pathways interact considerably as they project to V5/MT (Ferrera, Nealey, & Maunsell, 1992; Vidyasagar, Kulikowski, Lipnicki, & Dreher, 2002) (see Fig. 1), coherent motion may not uniquely isolate magnocellular mediated information. Therefore, despite being a dorsal stream measure, coherent motion cannot be considered a pure magnocellular measure. Alternatively, the spatial frequency doubling (FD) illusion is considered to be a low-level magnocellular measure (Kelly, 1966; Tyler, 1974). The FD illusion consists of coarse sinusoidal grating patterns, which, when modulated at high temporal and low spatial frequencies, creates the illusion of a stable grating with twice the actual spatial frequency of the component gratings. This apparent doubling was originally considered to be the result of nonlinear activity in retinal magnocellular cells (Bedford, Maddess, Rose, & James, 1997; Maddess et al., 1999; Tyler, 1974; White, Sun, Swanson, & Lee, 2002), however, more recent evidence indicates that the magnocellular pathway is isolated as a whole by the FD stimulus (Anderson & Johnson, 2002; White et al., 2002). Recent studies have used the FD illusion to examine magnocellular functioning in dyslexic readers, demonstrating that dyslexic children (Kevan & Pammer, 2008; Pammer & Kevan, 2007; Pammer & Vidyasagar, 2005; Pammer & Wheatley, 2001) and adults (Buchholz & McKone, 2004) are less sensitive to detecting the FD illusion than normal readers. The present study examines coherent motion and FD illusion sensitivity in two groups of children before they commence formal reading instruction; a selected group of children at familial risk for dyslexia and group of children unselected for family reading history. If there is a pre-existing dorsal stream deficit in children at-risk for dyslexia there is a simple and clear prediction – that at-risk children will demonstrate reduced sensitivity to detecting coherent motion and the FD illusion, whilst demonstrating normal performance in detecting non-dorsal control stimuli (coherent form and static gratings). 2. Methods 2.1. Participants Forty-two pre-reading children unselected for parental reading history (M = 5 years, 5 months, SD = 3 months; 23 boys, 19 girls) 1 In follow-up testing a year later, the groups were retrospectively reclassified based on the children’s Grade 1 literacy skills. When the data was reanalyzed, the groups differed significantly, whereby children with poor literacy skills in Grade 1 were found to have elevated coherent motion thresholds in pre-school (Boets, 2006).
and 20 pre-reading children at familial risk for dyslexia (M = 5 years, 6 months, SD = 7 months; 13 boys, 7 girls) participated in the study. The unselected children were recruited from local schools and the at-risk children were recruited through media announcements seeking children who were to enter kindergarten who had a first-degree family member with dyslexia. All children had normal or normal-to-corrected visual acuity and were of normal intelligence as assessed by the Brief Intellectual Ability measure of the Woodcock-Johnson III: Cognitive (Woodcock, McGrew, & Mather, 2001). The Letter–Word Identification subtest of the Woodcock-Johnson III: Achievement (Woodcock et al., 2001) was used to measure children’s emerging reading skills. 2.2. Stimuli and procedure 2.2.1. Coherent motion Motion coherence thresholds were measured using a random dot kinematogram consisting of a patch of 100 white dots (0.1o) randomly distributed within a 23 23o region on a black background. A variable proportion of these dots moved coherently, at a velocity of 4.4 degr/s, either upwards or downwards amongst the remaining randomly moving noise dots (Fig. 2). The number of coherently moving dots was manipulated according to a modified binary search (MOBS) threshold strategy (Tyrrell & Owens, 1988). On each trial the child indicated direction of motion. Threshold was defined as the lowest number of coherently moving dots required to perceive motion direction. Stimuli were presented as 18-frame sequences, with each frame lasting 16.7 ms. Both the signal and noise dots were randomly chosen on each animation frame. 2.2.2. Coherent form Coherent form stimuli consisted of a static array of 1024 randomly oriented white line segments presented within a 32 22.5o patch on a black background for 1800 ms (Fig. 2). The task was to indicate whether the target stimulus, a 8o region defined by lines oriented tangentially to concentric circles, was presented at the top or the bottom of the display. Threshold was defined as the proportion of coherently oriented line segments required to detect the circle target. 2.2.3. Frequency doubled (FD) gratings The FD stimuli were adapted from FDT perimeter technology (Welch Allyn, Skaneateles Falls, NY, and Carl Zeiss Meditec Inc., Dublin, CA), consisting of low spatial frequency vertical sinewave gratings presented within a square aperture (10o diameter) on a grey background. The 0.25 C/deg gratings were modulated at 50 Hz counterphase flicker, to create a percept of a stable grating with twice the actual spatial frequency of the component gratings (Fig. 3). At a viewing distance of approximately 40 cm, each stim-
2837
A. Kevan, K. Pammer / Vision Research 48 (2008) 2835–2839
Fig. 2. Schematic representations of the coherent motion (A, B) and coherent form (C, D) stimuli. For the motion task, arrows represent the motion vector of each dot during a given frame. Where (A) depicts 100% coherence and (B) depicts random motion (0%). For the form task, line segments are oriented to form a circle, where (C) represents 100% coherence and (D) depicts randomly oriented lines (0%). Note that illustrations are not drawn to scale, and that actual stimuli did not include distinct borders.
A
B
D
C
Fig. 3. The physical stimulus A is alternated with stimulus B. The resulting visual illusion is C, the FD stimulus. Stimulus D is the Fixed stimulus, which was designed to look like the FD illusion (C).
ulus subtended 10 10o of visual angle. On each trial a stimulus was presented 5o either to the left or to right of fixation for 720 ms, including a 160 ms ramped onset and offset. Children’s task was to indicate which side they saw the grating. Contrast threshold were again determined using MOBS, consisting of six staircase reversal and a test range of less than 6 dB. The range of possible threshold level values was between 0 dB (100%) maximum contrast (lowest sensitivity) and 40 dB (0%) minimum contrast (highest sensitivity). 2.2.4. Fixed gratings The Fixed condition employed the same procedure as the FD condition, differing only in the type of gratings. The Fixed gratings consisted of 0.5 C/deg static gratings that were not the result of the illusion, they were engineered to look like the product of the illusion, to allow for comparisons to be made between the FD and Fixed conditions. 2.3. General procedure To make the visual tests more child friendly each task had an associated animation which transformed a somewhat boring test into a computer game that the children wanted to ‘play’, e.g., coherent motion, dots moving coherently up and down the screen, was in the context of ‘helping the farmer catch the sheep’; coherent form, randomly oriented lines with an embedded circle, was in the context of ‘helping a girl find her ball hiding in the grass’, and the FD and Fixed conditions (striped patterns), were in the context of catching the zebra who was hiding to the left or the right of the screen. In each condition a two-alternative forced-choice procedure was used to estimate threshold. Children were required to indicate their response by pointing to the screen or by responding verbally. To ensure that the participants understood the tasks, each child completed a series of practice trials before commencing the test trials. All visual stimuli were presented on a BenQ 19 inch color monitor, driven by an Acer NVIDIA graphics card, with a screen refresh rate of 100 Hz.
3. Results Children were excluded from the analysis if their Letter–Word Identification (Woodcock et al., 2001) raw score was above 17, indicating a small sight vocabulary. The unselected group had a higher Letter–Word Identification score than the at-risk group, t(60) = 3.638, p < .01, d = .94. There were no significant age (p = .651) or intelligence (p = .695) differences between the groups (Table 1). Despite the two groups showing no significant difference in age or intelligence, we analyzed group data using Analysis of Covariance (ANCOVA) to control for the possible influence of these variables. Results indicate that the dorsal stream tasks significantly differentiate between the at-risk and unselected group. The at-risk children performed more poorly than the unselected children on both of the dorsal stream tasks whilst they performed as well as the unselected children on the control measures. The at-risk children were significantly less sensitive than the unselected children at detecting coherent motion, F(1, 59) = 4.762, p < .05, g2 = .55, whilst demonstrating no significant difference in detecting coherent form (p = .862). The dissociation was also demonstrated in the FD condition, detection thresholds for seeing the FD stimuli were significantly higher in the at-risk group than the unselected group, F(1, 59) = 8.753, p < .01, g2 = .78, whereas detection thresholds for the Fixed stimuli did not differ significantly between groups (p = .884,) (see Fig. 4).
Table 1 Descriptive statistics for at-risk and unselected children Measures
IQ Letter–Word ID Age (months)
At-risk (N = 20)
Unselected (N = 42)
Mean
SD
Mean
SD
107.40 89.2 67.50
9.41 10.14 9.17
111.47 98.76 66.52
11.13 8.29 3.71
2838
A. Kevan, K. Pammer / Vision Research 48 (2008) 2835–2839
Fig. 4. Mean contrast thresholds (±SEM) for the unselected and at-risk children in seeing: (A) Fixed and FD stimuli (lower scores (dB) indicate poorer performance); and (B) Coherent form and Coherent motion (higher scores (%) indicate poorer performance).
To determine whether the visual measures were valuable in classifying children as being at-risk or from the unselected sample, we performed a discriminant analysis using children’s coherent motion and FD illusion thresholds to predict group membership. A single discriminant analysis was calculated, with coherent motion and FD illusion thresholds entered simultaneously, v2(2) = 10.277, p = .01. Overall, the discriminant function correctly classified 71.4% of children as at-risk or unselected based on their scores, resulting in correct classification of 29 out of 42 normally developing children (69%) and 16 out of 21 (76.2%) at-risk children. To examine whether dorsal stream functioning differed in children with and without letter and word knowledge, children were separated into two groups based on Letter–Word Identification scores. Children more than one standard deviation (SD = 10) below the mean (M = 100) were considered to have poor letter–word knowledge (n = 19). Controlling for age and intelligence, results indicate significant group differences whereby children with poor letter and word knowledge demonstrated lowered sensitivity to seeing both coherent motion, F (1, 59) = 4.779, p < .05, g2 = .07, and FD sensitivity, F (1, 59) = 9.741, p < .01, g2 = .14 . 4. Discussion Our findings indicate that children who are at family risk of developing dyslexia possess a subtle visual impairment that is specific to the dorsal pathway before they start formal reading instruction. The at-risk children were significantly less sensitive to seeing the coherent motion compared to the unselected sample and this impairment appears to be specific to the dorsal stream as the at-risk group demonstrated no deficits in coherent form detection. The same dissociation was observed using the FD illusion and Fixed stimuli. The at-risk children were less sensitive to detecting the FD illusion compared to the unselected group, whilst both groups showed no difference to the Fixed grating control stimuli. These results support our prediction that dorsal stream impairments in children at familial risk for dyslexia exist prior to the commencement of formal reading instruction, suggesting that normally developing visual sensitivity is likely to be vital to the normal acquisition of reading skill. The children in this study had received no formal reading instruction, however there was evidence of letter knowledge and basic word recognition (e.g., ‘‘dog”, ‘‘as”) in some children. Therefore it may be useful for future research to consider using younger children who have no letter or word knowledge. It is essential that
such studies use age appropriate visual tasks, as our pilot testing indicated that coherent motion detection tasks would be too difficult for children much younger than the age range used here. Conversely, although the FD illusion has not been used with younger children, experiential evidence suggests that the FD illusion may be a suitable test for younger children, though further investigation is required. One of the big remaining questions in this literature is a plausible argument outlining how a dorsal stream deficit might lead to reading difficulties. One possibility is that the dorsal stream may be involved in reading by virtue of its role in pre-attentive spatial coding (Pammer & Vidyasagar, 2005; Vidyasagar, 1999). Thus reading difficulties may arise if the dorsal stream does not provide adequate resources to guide saccadic eye movements and maintain stable fixations, which in turn could be a consequence of poor oculomotor control resulting from inadequate attentional feedback from the dorsal stream. Moreover, inadequate spatial sampling by the dorsal stream may result in localized deficits in the coding of initial stimulus features that would normally be fed back to the ventral stream for feature binding. Such problems could result in impaired orthographic skills as it would be difficult to develop stable lexical entries i.e., poor spatial sampling might confuse local feature elements in a way that would make letter position encoding difficult. Here we have demonstrated that visual deficits occur before reading commencement – the causal consequence of these deficits in dyslexia remains open to speculation. Certainly, these results demonstrate that visual deficits are not the consequence of failing to learn to read. However, future research should explore whether dorsal stream deficits play a causal role in reading failure, or whether they represent a biological marker that is associated with more general cognitive impairments. 5. Conclusion Reading involves the synthesis of a visual pattern with a language code, and dyslexia has traditionally been considered to be a deficit in the utilization of the language code. The present results however indicate that children at-risk for dyslexia have an impairment in the visual coding mechanisms before they learn to read. Understanding the contribution of visual mechanisms underlying reading is vital to the development of comprehensive assessments for identifying impairment in the cognitive skills required for proficient reading, and is crucial to the development of effective remediation.
A. Kevan, K. Pammer / Vision Research 48 (2008) 2835–2839
Acknowledgments We are grateful to the children, parents, teachers and schools for their willing participation and to Andy Thomson for his assistance in writing the programs for these experiments. References Anderson, A. J., & Johnson, C. A. (2002). Mechanisms isolated by frequency-doubling technology perimetry. Investigative Ophthalmology and Visual Sciences, 43(2), 398–401. Bedford, S., Maddess, T., Rose, K. A., & James, A. C. (1997). Correlations between observability of the spatial frequency doubled illusion and a multi-region pattern electroretinogram. Australian and New Zealand Journal of Ophthalmology, 25, S91–S93. Boden, C., & Giaschi, D. (2007). M-stream deficits and reading-related visual processes in developmental dyslexia. Psychological Bulletin, 133(2), 346–366. Boets, B. (2006). Early literacy development in children at risk for dyslexia. A longitudinal study of the general magnocellular theory. Unpublished doctoral dissertation, Katholieke Universiteit Leuven, Leuven. Boets, B., Wounters, J., van Wieringen, A., & Ghesquière, P. (2006). Coherent motion in preschool children at family risk for dyslexia. Vision Research, 46, 527–535. Buchholz, J., & McKone, E. (2004). Adults with dyslexia show deficits on spatial frequency doubling and visual attention tasks. Dyslexia, 10, 24–43. Cornelissen, P. L., Richardson, A., Mason, A., Fowler, S., & Stein, J. (1995). Contrast sensitivity and coherent motion detection measured at photopic luminance levels in dyslexics and controls. Vision Research, 35(10), 1483–1494. Demb, J. B., Boynton, G. M., Best, M., & Heeger, D. J. (1998a). Psychophysical evidence for a magnocellular pathway deficit in dyslexia. Vision Research, 38, 1555–1559. Demb, J. B., Boynton, G. M., & Heeger, D. J. (1998b). Functional magnetic resonance imaging of early visual pathways in dyslexia. The Journal of Neuroscience, 18(17), 6939–6951. Eden, G. F., VanMeter, J. W., Rumsey, J. M., Maisog, J. M., Woods, R. P., & Zeffiro, T. A. (1996). Abnormal processing of visual motion in dyslexia revealed by functional brain imaging. Nature, 382, 66–69. Ferrera, V. P., Nealey, T. A., & Maunsell, J. H. R. (1992). Mixed parvocellular and magnocellular geniculate signals in visual area V4. Nature, 358, 756–761. Galaburda, A. M., & Livingstone, M. S. (1993). Physiological evidence for a magnocellular defect in developmental dyslexia. Annals New York Academy of Sciences, 682, 70–82. Hulslander, J., Talcott, J. B., Witton, C., DeFries, J., Pennington, B. F., Wadsworth, S., et al. (2004). Sensory processing, reading, IQ, and attention. Journal of Experimental Child Psychology, 88, 274–295. Kelly, D. H. (1966). Frequency doubling in visual responses. Journal of the Optical Society of America, 56(11), 1628–1633. Kevan, A., & Pammer, K. (2008). Making the link between dorsal stream sensitivity and reading. Neuroreport, 19(4), 467–470. Kronbichler, M., Hutzler, F., & Wimmer, H. (2002). Dyslexia: Verbal impairments in the absence of magnocellular impairments. Neuroreport, 13(5), 617–620. Livingstone, M. S., Rosen, G. D., Drislane, F. W., & Galaburda, A. M. (1991). Physiological and anatomical evidence for a magnocellular defect in
2839
developmental dyslexia. Proceedings of the National Academy of Sciences of the USA, 88, 7943–7947. Maddess, T., Goldberg, I., Dobinson, J., Wine, S., Welsh, A. H., & James, A. C. (1999). Testing for glaucoma with the spatial frequency doubling illusion. Vision Research, 39, 4258–4273. Merigan, W. H., & Maunsell, J. H. R. (1993). How parallel are the primate visual pathways? Annual Review of Neuroscience, 16, 369–402. Pammer, K., & Kevan, A. (2007). The contribution of visual sensitivity, phonological processing and non-verbal IQ to children’s reading. Scientific Studies of Reading, 11(1), 33–53. Pammer, K., & Vidyasagar, T. R. (2005). Integration of the visual and auditory networks in dyslexia: a theoretical perspective. Journal of Research in Reading, 28(3), 320–331. Pammer, K., & Wheatley, C. (2001). Isolating the M(y)-cell response in dyslexia using the spatial frequency doubling illusion. Vision Research, 41, 2139–2147. Ramus, F. (2001). Dyslexia: Talk of two theories. Nature, 412, 393–395. Ramus, F. (2003). Developmental dyslexia: Specific phonological deficit or general sensorimotor dysfunction? Current Opinion in Neurobiology, 13, 212–218. Skottun, B. C. (2000). On the conflicting support for the magnocellular-deficit theory of dyslexia. Trends in Cognitive Sciences, 4(6), 211–212. Skottun, B. C., & Skoyles, J. R. (2006). Is coherent motion an appropriate test for magnocellular sensitivity? Brain and Cognition, 61, 172–180. Stein, J. F. (2001). The magnocellular theory of developmental dyslexia. Dyslexia, 7, 12–36. Talcott, J. B., Hansen, P. C., Assoku, E. L., & Stein, J. F. (2000). Visual motion sensitivity in dyslexia: evidence for temporal and energy integration deficits. Neuropsychologia, 38, 935–943. Tyler, C. W. (1974). Observations on spatial-frequency doubling. Perception, 3, 81–86. Tyrrell, R. A., & Owens, D. A. (1988). A rapid technique to assess the resting states of the eyes and other threshold phenomena: The modified binary search (MOBS). Behavior Research Methods, Instruments & Computers, 20(2), 137–141. Vellutino, F. R., Fletcher, J. M., Snowling, M. J., & Scanlon, D. M. (2004). Specific reading disability (dyslexia): What have we learned in the past four decades? Journal of Child Psychology and Psychiatry, 45(1), 2–40. Vidyasagar, T. R. (1999). A neuronal model of attentional spotlight: parietal guiding the temporal. Brain Research Reviews, 30, 66–76. Vidyasagar, T. R. (2004). Neural underpinnings of dyslexia as a disorder of visuospatial attention. Clinical and Experimental Optometry, 87(1), 4–10. Vidyasagar, T. R., Kulikowski, J. J., Lipnicki, D. M., & Dreher, B. (2002). Convergence of parvocellular and magnocellular information channels in the primary visual cortex of the macaque. European Journal of Neuroscience, 16, 945–956. White, A. J. R., Sun, H., Swanson, W. H., & Lee, B. B. (2002). An examination of physiological mechanisms underlying the frequency-doubling illusion. Investigative Ophthalmology and Visual Sciences, 43(11), 3590–3599. Whitney, C., & Cornelissen, P. L. (2005). Letter position encoding and dyslexia. Journal of Research in Reading, 28, 274–301. Wilmer, J. B., Richardson, A. J., Chen, Y., & Stein, J. F. (2004). Two visual motion processing deficits in developmental dyslexia associated with different reading skills deficits. Journal of Cognitive Neuroscience, 16(4), 528–540. Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock–Johnson III tests of: Achievement and cognitive abilities (3rd ed.). Itasca, Illinois: Riverside Publishing.
Contents lists available at ScienceDirect
Vision Research journal homepage: www.elsevier.com/locate/visres
Visual deficits in pre-readers at familial risk for dyslexia Alison Kevan *, Kristen Pammer The Australian National University, School of Psychology, Canberra ACT 0200, Australia
a r t i c l e
i n f o
Article history: Received 6 August 2008 Received in revised form 26 September 2008
Keywords: Dorsal stream Magnocellular Dyslexia Visual frequency doubling Coherent motion Pre-reading children
a b s t r a c t Visual processing deficits in dyslexic readers are argued to evolve as a consequence of reading failure. This study examines dorsal stream functioning of children before they commence formal reading instruction to determine whether visual deficits precede reading difficulties. Coherent motion and visual frequency doubling detection were measured in children at familial risk for dyslexia and in children unselected for family reading history. Here we show that children who are at family risk for dyslexia demonstrate dorsal stream deficits before they learn to read, whilst demonstrating no corresponding deficits in coherent form and static grating control tasks. Results indicate that the dorsal visual deficits observed in dyslexic readers are unlikely to be the result of reading failure. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction The most widely acknowledged impairment in dyslexia is a difficulty in phonological processing. As such dyslexia is fundamentally considered to be a language based problem in the cognitive manipulation of words (Vellutino, Fletcher, Snowling, & Scanlon, 2004). However, it is now clear that a phonological account on its own cannot explain the full extent of impairments observed in dyslexic readers. Thirty years of research indicates that many dyslexic readers also possess a visual processing deficit that is specific to the magnocellular dominated dorsal pathway. Convergent evidence from behavioral (e.g., Cornelissen, Richardson, Mason, Fowler, & Stein, 1995; Kevan & Pammer, 2008; Pammer & Kevan, 2007; Pammer & Vidyasagar, 2005; Pammer & Wheatley, 2001; Talcott, Hansen, Assoku, & Stein, 2000; Wilmer, Richardson, Chen, & Stein, 2004), anatomical (e.g., Galaburda & Livingstone, 1993; Livingstone, Rosen, Drislane, & Galaburda, 1991) and imaging studies (e.g., Demb, Boynton, Best, & Heeger, 1998a; Demb, Boynton, & Heeger, 1998b; Eden et al., 1996) provide support to suggest that dyslexic readers are less sensitive to visual stimuli that is mediated by the dorsal stream. These findings have led to theoretical models that explain the link between a dorsal stream deficit and reading, including attention (Vidyasagar, 2004), ocular motor control (Stein, 2001), and letter position encoding (Whitney & Cornelissen, 2005), see Boden and Giaschi (2007) for a recent review. The suggestion of a dorsal stream deficit in dyslexic readers that is related to reading failure however is controversial (Ramus, 2001,
* Corresponding author. Fax: +61 2 6125 0499. E-mail address: [email protected] (A. Kevan). 0042-6989/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.visres.2008.09.022
2003; Skottun, 2000). In fact, it is counter-intuitive that the dorsal (visual ‘‘where”) stream should have a role in reading as the mechanisms for the detailed feature analysis required for reading should lie with the ventral (visual ‘‘what”) stream, which has a higher spatial resolution and a role in object recognition. Thus visual deficits are frequently dismissed as being secondary to a core language deficit, where dorsal stream impairments develop as a consequence of failing to learn to read (Ramus, 2003; Vellutino et al., 2004), as the reading network does not receive adequate visual exposure from print and thus may not provide sufficient input to the system to stimulate neural connections. Indeed, it is biologically plausible that such reciprocal feedback could occur. The reading network is highly interactive and dynamic, requiring the synthesis of signals from disparate areas of the brain, and learning to read requires integrating and tuning different components of that network in a highly sophisticated way. Thus, Hebbian learning principles would predict that impairment in one part of the network (e.g. a core language deficit) could result in failure to develop other components of the network, (e.g., visuo-spatial encoding). Therefore it is currently unknown whether the dorsal stream deficits are a cause or a consequence of reading failure. However, if it could be demonstrated that dorsal stream deficits exist in children at family risk of dyslexia prior to the onset of reading, this would challenge the argument that dorsal stream deficits are the result of reading failure. Numerous studies have demonstrated that dyslexic readers are less sensitive to coherent motion than control readers (Cornelissen et al., 1995; Hulslander et al., 2004; Pammer & Wheatley, 2001; Talcott et al., 2000; Wilmer et al., 2004). This is in contrast to normal performance on static tasks measuring coherent form detection (Hulslander et al., 2004; Wilmer et al., 2004). This dissociation suggests that dyslexic readers possess an impairment specific to the
2836
A. Kevan, K. Pammer / Vision Research 48 (2008) 2835–2839
Magno Pathway
Parvo Pathway
Retina
LGN
M cells
Magno
Layer 4Cα
Parvo
Layer 4C β
P cells
V1
V2 Layer 4β
Thick Stripes
Blob
Thin Stripes
InterBlobs
Extra striate
InterStripes
V5/MT
Dorsal Stream
V4
Ventral Stream
Fig. 1. Representation of the parallel pathways in the primate visual system. Lines indicate connections between components. From V1 magno and parvocellular information interact considerably as they project to extrastriate visual areas (Merigan & Maunsell, 1993; Vidyasagar et al., 2002), suggesting that higher order dorsal processing may not be entirely indicative of lower level magno functioning. LGN: lateral geniculate nucleus, MT: middle temporal area. Adapted from Merigan and Maunsell (1993).
detection of the dynamic properties of visual stimuli, and adds further support to the notion of a dorsal stream deficit. However, support for the use of coherent motion is not universal (Skottun & Skoyles, 2006), as equivocal results have been demonstrated in dyslexic children (Kronbichler, Hutzler, & Wimmer, 2002) and in prereaders1 (Boets, Wounters, van Wieringen, & Ghesquière, 2006). Given that magnocellular and parvocellular pathways interact considerably as they project to V5/MT (Ferrera, Nealey, & Maunsell, 1992; Vidyasagar, Kulikowski, Lipnicki, & Dreher, 2002) (see Fig. 1), coherent motion may not uniquely isolate magnocellular mediated information. Therefore, despite being a dorsal stream measure, coherent motion cannot be considered a pure magnocellular measure. Alternatively, the spatial frequency doubling (FD) illusion is considered to be a low-level magnocellular measure (Kelly, 1966; Tyler, 1974). The FD illusion consists of coarse sinusoidal grating patterns, which, when modulated at high temporal and low spatial frequencies, creates the illusion of a stable grating with twice the actual spatial frequency of the component gratings. This apparent doubling was originally considered to be the result of nonlinear activity in retinal magnocellular cells (Bedford, Maddess, Rose, & James, 1997; Maddess et al., 1999; Tyler, 1974; White, Sun, Swanson, & Lee, 2002), however, more recent evidence indicates that the magnocellular pathway is isolated as a whole by the FD stimulus (Anderson & Johnson, 2002; White et al., 2002). Recent studies have used the FD illusion to examine magnocellular functioning in dyslexic readers, demonstrating that dyslexic children (Kevan & Pammer, 2008; Pammer & Kevan, 2007; Pammer & Vidyasagar, 2005; Pammer & Wheatley, 2001) and adults (Buchholz & McKone, 2004) are less sensitive to detecting the FD illusion than normal readers. The present study examines coherent motion and FD illusion sensitivity in two groups of children before they commence formal reading instruction; a selected group of children at familial risk for dyslexia and group of children unselected for family reading history. If there is a pre-existing dorsal stream deficit in children at-risk for dyslexia there is a simple and clear prediction – that at-risk children will demonstrate reduced sensitivity to detecting coherent motion and the FD illusion, whilst demonstrating normal performance in detecting non-dorsal control stimuli (coherent form and static gratings). 2. Methods 2.1. Participants Forty-two pre-reading children unselected for parental reading history (M = 5 years, 5 months, SD = 3 months; 23 boys, 19 girls) 1 In follow-up testing a year later, the groups were retrospectively reclassified based on the children’s Grade 1 literacy skills. When the data was reanalyzed, the groups differed significantly, whereby children with poor literacy skills in Grade 1 were found to have elevated coherent motion thresholds in pre-school (Boets, 2006).
and 20 pre-reading children at familial risk for dyslexia (M = 5 years, 6 months, SD = 7 months; 13 boys, 7 girls) participated in the study. The unselected children were recruited from local schools and the at-risk children were recruited through media announcements seeking children who were to enter kindergarten who had a first-degree family member with dyslexia. All children had normal or normal-to-corrected visual acuity and were of normal intelligence as assessed by the Brief Intellectual Ability measure of the Woodcock-Johnson III: Cognitive (Woodcock, McGrew, & Mather, 2001). The Letter–Word Identification subtest of the Woodcock-Johnson III: Achievement (Woodcock et al., 2001) was used to measure children’s emerging reading skills. 2.2. Stimuli and procedure 2.2.1. Coherent motion Motion coherence thresholds were measured using a random dot kinematogram consisting of a patch of 100 white dots (0.1o) randomly distributed within a 23 23o region on a black background. A variable proportion of these dots moved coherently, at a velocity of 4.4 degr/s, either upwards or downwards amongst the remaining randomly moving noise dots (Fig. 2). The number of coherently moving dots was manipulated according to a modified binary search (MOBS) threshold strategy (Tyrrell & Owens, 1988). On each trial the child indicated direction of motion. Threshold was defined as the lowest number of coherently moving dots required to perceive motion direction. Stimuli were presented as 18-frame sequences, with each frame lasting 16.7 ms. Both the signal and noise dots were randomly chosen on each animation frame. 2.2.2. Coherent form Coherent form stimuli consisted of a static array of 1024 randomly oriented white line segments presented within a 32 22.5o patch on a black background for 1800 ms (Fig. 2). The task was to indicate whether the target stimulus, a 8o region defined by lines oriented tangentially to concentric circles, was presented at the top or the bottom of the display. Threshold was defined as the proportion of coherently oriented line segments required to detect the circle target. 2.2.3. Frequency doubled (FD) gratings The FD stimuli were adapted from FDT perimeter technology (Welch Allyn, Skaneateles Falls, NY, and Carl Zeiss Meditec Inc., Dublin, CA), consisting of low spatial frequency vertical sinewave gratings presented within a square aperture (10o diameter) on a grey background. The 0.25 C/deg gratings were modulated at 50 Hz counterphase flicker, to create a percept of a stable grating with twice the actual spatial frequency of the component gratings (Fig. 3). At a viewing distance of approximately 40 cm, each stim-
2837
A. Kevan, K. Pammer / Vision Research 48 (2008) 2835–2839
Fig. 2. Schematic representations of the coherent motion (A, B) and coherent form (C, D) stimuli. For the motion task, arrows represent the motion vector of each dot during a given frame. Where (A) depicts 100% coherence and (B) depicts random motion (0%). For the form task, line segments are oriented to form a circle, where (C) represents 100% coherence and (D) depicts randomly oriented lines (0%). Note that illustrations are not drawn to scale, and that actual stimuli did not include distinct borders.
A
B
D
C
Fig. 3. The physical stimulus A is alternated with stimulus B. The resulting visual illusion is C, the FD stimulus. Stimulus D is the Fixed stimulus, which was designed to look like the FD illusion (C).
ulus subtended 10 10o of visual angle. On each trial a stimulus was presented 5o either to the left or to right of fixation for 720 ms, including a 160 ms ramped onset and offset. Children’s task was to indicate which side they saw the grating. Contrast threshold were again determined using MOBS, consisting of six staircase reversal and a test range of less than 6 dB. The range of possible threshold level values was between 0 dB (100%) maximum contrast (lowest sensitivity) and 40 dB (0%) minimum contrast (highest sensitivity). 2.2.4. Fixed gratings The Fixed condition employed the same procedure as the FD condition, differing only in the type of gratings. The Fixed gratings consisted of 0.5 C/deg static gratings that were not the result of the illusion, they were engineered to look like the product of the illusion, to allow for comparisons to be made between the FD and Fixed conditions. 2.3. General procedure To make the visual tests more child friendly each task had an associated animation which transformed a somewhat boring test into a computer game that the children wanted to ‘play’, e.g., coherent motion, dots moving coherently up and down the screen, was in the context of ‘helping the farmer catch the sheep’; coherent form, randomly oriented lines with an embedded circle, was in the context of ‘helping a girl find her ball hiding in the grass’, and the FD and Fixed conditions (striped patterns), were in the context of catching the zebra who was hiding to the left or the right of the screen. In each condition a two-alternative forced-choice procedure was used to estimate threshold. Children were required to indicate their response by pointing to the screen or by responding verbally. To ensure that the participants understood the tasks, each child completed a series of practice trials before commencing the test trials. All visual stimuli were presented on a BenQ 19 inch color monitor, driven by an Acer NVIDIA graphics card, with a screen refresh rate of 100 Hz.
3. Results Children were excluded from the analysis if their Letter–Word Identification (Woodcock et al., 2001) raw score was above 17, indicating a small sight vocabulary. The unselected group had a higher Letter–Word Identification score than the at-risk group, t(60) = 3.638, p < .01, d = .94. There were no significant age (p = .651) or intelligence (p = .695) differences between the groups (Table 1). Despite the two groups showing no significant difference in age or intelligence, we analyzed group data using Analysis of Covariance (ANCOVA) to control for the possible influence of these variables. Results indicate that the dorsal stream tasks significantly differentiate between the at-risk and unselected group. The at-risk children performed more poorly than the unselected children on both of the dorsal stream tasks whilst they performed as well as the unselected children on the control measures. The at-risk children were significantly less sensitive than the unselected children at detecting coherent motion, F(1, 59) = 4.762, p < .05, g2 = .55, whilst demonstrating no significant difference in detecting coherent form (p = .862). The dissociation was also demonstrated in the FD condition, detection thresholds for seeing the FD stimuli were significantly higher in the at-risk group than the unselected group, F(1, 59) = 8.753, p < .01, g2 = .78, whereas detection thresholds for the Fixed stimuli did not differ significantly between groups (p = .884,) (see Fig. 4).
Table 1 Descriptive statistics for at-risk and unselected children Measures
IQ Letter–Word ID Age (months)
At-risk (N = 20)
Unselected (N = 42)
Mean
SD
Mean
SD
107.40 89.2 67.50
9.41 10.14 9.17
111.47 98.76 66.52
11.13 8.29 3.71
2838
A. Kevan, K. Pammer / Vision Research 48 (2008) 2835–2839
Fig. 4. Mean contrast thresholds (±SEM) for the unselected and at-risk children in seeing: (A) Fixed and FD stimuli (lower scores (dB) indicate poorer performance); and (B) Coherent form and Coherent motion (higher scores (%) indicate poorer performance).
To determine whether the visual measures were valuable in classifying children as being at-risk or from the unselected sample, we performed a discriminant analysis using children’s coherent motion and FD illusion thresholds to predict group membership. A single discriminant analysis was calculated, with coherent motion and FD illusion thresholds entered simultaneously, v2(2) = 10.277, p = .01. Overall, the discriminant function correctly classified 71.4% of children as at-risk or unselected based on their scores, resulting in correct classification of 29 out of 42 normally developing children (69%) and 16 out of 21 (76.2%) at-risk children. To examine whether dorsal stream functioning differed in children with and without letter and word knowledge, children were separated into two groups based on Letter–Word Identification scores. Children more than one standard deviation (SD = 10) below the mean (M = 100) were considered to have poor letter–word knowledge (n = 19). Controlling for age and intelligence, results indicate significant group differences whereby children with poor letter and word knowledge demonstrated lowered sensitivity to seeing both coherent motion, F (1, 59) = 4.779, p < .05, g2 = .07, and FD sensitivity, F (1, 59) = 9.741, p < .01, g2 = .14 . 4. Discussion Our findings indicate that children who are at family risk of developing dyslexia possess a subtle visual impairment that is specific to the dorsal pathway before they start formal reading instruction. The at-risk children were significantly less sensitive to seeing the coherent motion compared to the unselected sample and this impairment appears to be specific to the dorsal stream as the at-risk group demonstrated no deficits in coherent form detection. The same dissociation was observed using the FD illusion and Fixed stimuli. The at-risk children were less sensitive to detecting the FD illusion compared to the unselected group, whilst both groups showed no difference to the Fixed grating control stimuli. These results support our prediction that dorsal stream impairments in children at familial risk for dyslexia exist prior to the commencement of formal reading instruction, suggesting that normally developing visual sensitivity is likely to be vital to the normal acquisition of reading skill. The children in this study had received no formal reading instruction, however there was evidence of letter knowledge and basic word recognition (e.g., ‘‘dog”, ‘‘as”) in some children. Therefore it may be useful for future research to consider using younger children who have no letter or word knowledge. It is essential that
such studies use age appropriate visual tasks, as our pilot testing indicated that coherent motion detection tasks would be too difficult for children much younger than the age range used here. Conversely, although the FD illusion has not been used with younger children, experiential evidence suggests that the FD illusion may be a suitable test for younger children, though further investigation is required. One of the big remaining questions in this literature is a plausible argument outlining how a dorsal stream deficit might lead to reading difficulties. One possibility is that the dorsal stream may be involved in reading by virtue of its role in pre-attentive spatial coding (Pammer & Vidyasagar, 2005; Vidyasagar, 1999). Thus reading difficulties may arise if the dorsal stream does not provide adequate resources to guide saccadic eye movements and maintain stable fixations, which in turn could be a consequence of poor oculomotor control resulting from inadequate attentional feedback from the dorsal stream. Moreover, inadequate spatial sampling by the dorsal stream may result in localized deficits in the coding of initial stimulus features that would normally be fed back to the ventral stream for feature binding. Such problems could result in impaired orthographic skills as it would be difficult to develop stable lexical entries i.e., poor spatial sampling might confuse local feature elements in a way that would make letter position encoding difficult. Here we have demonstrated that visual deficits occur before reading commencement – the causal consequence of these deficits in dyslexia remains open to speculation. Certainly, these results demonstrate that visual deficits are not the consequence of failing to learn to read. However, future research should explore whether dorsal stream deficits play a causal role in reading failure, or whether they represent a biological marker that is associated with more general cognitive impairments. 5. Conclusion Reading involves the synthesis of a visual pattern with a language code, and dyslexia has traditionally been considered to be a deficit in the utilization of the language code. The present results however indicate that children at-risk for dyslexia have an impairment in the visual coding mechanisms before they learn to read. Understanding the contribution of visual mechanisms underlying reading is vital to the development of comprehensive assessments for identifying impairment in the cognitive skills required for proficient reading, and is crucial to the development of effective remediation.
A. Kevan, K. Pammer / Vision Research 48 (2008) 2835–2839
Acknowledgments We are grateful to the children, parents, teachers and schools for their willing participation and to Andy Thomson for his assistance in writing the programs for these experiments. References Anderson, A. J., & Johnson, C. A. (2002). Mechanisms isolated by frequency-doubling technology perimetry. Investigative Ophthalmology and Visual Sciences, 43(2), 398–401. Bedford, S., Maddess, T., Rose, K. A., & James, A. C. (1997). Correlations between observability of the spatial frequency doubled illusion and a multi-region pattern electroretinogram. Australian and New Zealand Journal of Ophthalmology, 25, S91–S93. Boden, C., & Giaschi, D. (2007). M-stream deficits and reading-related visual processes in developmental dyslexia. Psychological Bulletin, 133(2), 346–366. Boets, B. (2006). Early literacy development in children at risk for dyslexia. A longitudinal study of the general magnocellular theory. Unpublished doctoral dissertation, Katholieke Universiteit Leuven, Leuven. Boets, B., Wounters, J., van Wieringen, A., & Ghesquière, P. (2006). Coherent motion in preschool children at family risk for dyslexia. Vision Research, 46, 527–535. Buchholz, J., & McKone, E. (2004). Adults with dyslexia show deficits on spatial frequency doubling and visual attention tasks. Dyslexia, 10, 24–43. Cornelissen, P. L., Richardson, A., Mason, A., Fowler, S., & Stein, J. (1995). Contrast sensitivity and coherent motion detection measured at photopic luminance levels in dyslexics and controls. Vision Research, 35(10), 1483–1494. Demb, J. B., Boynton, G. M., Best, M., & Heeger, D. J. (1998a). Psychophysical evidence for a magnocellular pathway deficit in dyslexia. Vision Research, 38, 1555–1559. Demb, J. B., Boynton, G. M., & Heeger, D. J. (1998b). Functional magnetic resonance imaging of early visual pathways in dyslexia. The Journal of Neuroscience, 18(17), 6939–6951. Eden, G. F., VanMeter, J. W., Rumsey, J. M., Maisog, J. M., Woods, R. P., & Zeffiro, T. A. (1996). Abnormal processing of visual motion in dyslexia revealed by functional brain imaging. Nature, 382, 66–69. Ferrera, V. P., Nealey, T. A., & Maunsell, J. H. R. (1992). Mixed parvocellular and magnocellular geniculate signals in visual area V4. Nature, 358, 756–761. Galaburda, A. M., & Livingstone, M. S. (1993). Physiological evidence for a magnocellular defect in developmental dyslexia. Annals New York Academy of Sciences, 682, 70–82. Hulslander, J., Talcott, J. B., Witton, C., DeFries, J., Pennington, B. F., Wadsworth, S., et al. (2004). Sensory processing, reading, IQ, and attention. Journal of Experimental Child Psychology, 88, 274–295. Kelly, D. H. (1966). Frequency doubling in visual responses. Journal of the Optical Society of America, 56(11), 1628–1633. Kevan, A., & Pammer, K. (2008). Making the link between dorsal stream sensitivity and reading. Neuroreport, 19(4), 467–470. Kronbichler, M., Hutzler, F., & Wimmer, H. (2002). Dyslexia: Verbal impairments in the absence of magnocellular impairments. Neuroreport, 13(5), 617–620. Livingstone, M. S., Rosen, G. D., Drislane, F. W., & Galaburda, A. M. (1991). Physiological and anatomical evidence for a magnocellular defect in
2839
developmental dyslexia. Proceedings of the National Academy of Sciences of the USA, 88, 7943–7947. Maddess, T., Goldberg, I., Dobinson, J., Wine, S., Welsh, A. H., & James, A. C. (1999). Testing for glaucoma with the spatial frequency doubling illusion. Vision Research, 39, 4258–4273. Merigan, W. H., & Maunsell, J. H. R. (1993). How parallel are the primate visual pathways? Annual Review of Neuroscience, 16, 369–402. Pammer, K., & Kevan, A. (2007). The contribution of visual sensitivity, phonological processing and non-verbal IQ to children’s reading. Scientific Studies of Reading, 11(1), 33–53. Pammer, K., & Vidyasagar, T. R. (2005). Integration of the visual and auditory networks in dyslexia: a theoretical perspective. Journal of Research in Reading, 28(3), 320–331. Pammer, K., & Wheatley, C. (2001). Isolating the M(y)-cell response in dyslexia using the spatial frequency doubling illusion. Vision Research, 41, 2139–2147. Ramus, F. (2001). Dyslexia: Talk of two theories. Nature, 412, 393–395. Ramus, F. (2003). Developmental dyslexia: Specific phonological deficit or general sensorimotor dysfunction? Current Opinion in Neurobiology, 13, 212–218. Skottun, B. C. (2000). On the conflicting support for the magnocellular-deficit theory of dyslexia. Trends in Cognitive Sciences, 4(6), 211–212. Skottun, B. C., & Skoyles, J. R. (2006). Is coherent motion an appropriate test for magnocellular sensitivity? Brain and Cognition, 61, 172–180. Stein, J. F. (2001). The magnocellular theory of developmental dyslexia. Dyslexia, 7, 12–36. Talcott, J. B., Hansen, P. C., Assoku, E. L., & Stein, J. F. (2000). Visual motion sensitivity in dyslexia: evidence for temporal and energy integration deficits. Neuropsychologia, 38, 935–943. Tyler, C. W. (1974). Observations on spatial-frequency doubling. Perception, 3, 81–86. Tyrrell, R. A., & Owens, D. A. (1988). A rapid technique to assess the resting states of the eyes and other threshold phenomena: The modified binary search (MOBS). Behavior Research Methods, Instruments & Computers, 20(2), 137–141. Vellutino, F. R., Fletcher, J. M., Snowling, M. J., & Scanlon, D. M. (2004). Specific reading disability (dyslexia): What have we learned in the past four decades? Journal of Child Psychology and Psychiatry, 45(1), 2–40. Vidyasagar, T. R. (1999). A neuronal model of attentional spotlight: parietal guiding the temporal. Brain Research Reviews, 30, 66–76. Vidyasagar, T. R. (2004). Neural underpinnings of dyslexia as a disorder of visuospatial attention. Clinical and Experimental Optometry, 87(1), 4–10. Vidyasagar, T. R., Kulikowski, J. J., Lipnicki, D. M., & Dreher, B. (2002). Convergence of parvocellular and magnocellular information channels in the primary visual cortex of the macaque. European Journal of Neuroscience, 16, 945–956. White, A. J. R., Sun, H., Swanson, W. H., & Lee, B. B. (2002). An examination of physiological mechanisms underlying the frequency-doubling illusion. Investigative Ophthalmology and Visual Sciences, 43(11), 3590–3599. Whitney, C., & Cornelissen, P. L. (2005). Letter position encoding and dyslexia. Journal of Research in Reading, 28, 274–301. Wilmer, J. B., Richardson, A. J., Chen, Y., & Stein, J. F. (2004). Two visual motion processing deficits in developmental dyslexia associated with different reading skills deficits. Journal of Cognitive Neuroscience, 16(4), 528–540. Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock–Johnson III tests of: Achievement and cognitive abilities (3rd ed.). Itasca, Illinois: Riverside Publishing.