1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References


To examine visual sensory and perceptive functions, study their interrelations, and explore associations between visual dysfunctions and intelligence in very preterm/very-low-birthweight (VP/VLBW) children.


One-hundred and sixteen VP/VLBW children (57 males, 59 females; mean gestational age 30.1wks, SD 2.3; mean corrected age 5y 6mo, SD 1mo) and 73 term-born children (40 males, 33 females; mean gestational age 39.9wks, SD 1.3; mean age 5y 6mo, SD 3mo) completed visual sensory (acuity, visual field, contrast-, color-, and stereovision), perceptive (visual coherence, and Developmental Test of Visual Perception non-motor scale), and intelligence assessments.


Compared with term-born children, VP/VLBW children had reduced acuity (d=0.70, p<0.001), inferior visual field (d=0.67, p<0.001), and stereovision (v=0.19, p=0.008). VP/VBLW children showed weaker static coherence (d=0.49, p=0.001) and Position in Space (d=0.41, p=0.006) performance, independent of visual sensory deficits, and showed lower Verbal IQ (VIQ) and Performance IQ (PIQ; p<0.001). Visual perceptive functioning accounted for 13% of variance in VIQ, and for 35% of variance in PIQ.


Visual sensory and perceptive dysfunctions are present in VP/VLBW children and occur largely independently of each other. Visual perceptive dysfunctions are moderately associated with PIQ, and weakly with VIQ.


Developmental Test of Visual Perception, 2nd edition


Performance IQ


Retinopathy of prematurity


Verbal IQ




Very preterm/very-low-birthweight

Very preterm-born children are at risk of damage to both peripheral and central structures of the visual system. Retinopathy of prematurity (ROP) accounts for the highest rates of ocular morbidity in very preterm survivors, but visual dysfunctions are also present in very preterm children without ROP.[1] Abnormalities in brain development result in reduced brain volumes and disturbed integrity of cerebral white matter.[2] Consequently, cerebral visual impairments have been recognized as a major cause of visual impairment among preterm-born children[3] and may occur when the central structures of the visual system are affected.

Large cohort studies report visual sensory deficits in very preterm children, including ocular misalignment and reduced visual acuity, contrast sensitivity, and stereovision.[4, 5] In addition, a recent meta-analysis showed that visual perceptive dysfunctions are most prominent on tasks requiring visual-spatial analysis.[6] It is, however, unclear whether visual perceptive dysfunctions exist either independently or as a result of visual sensory deficits.

Visual perceptive abilities have recently been identified as one of the neurocognitive factors underlying[7] the well established IQ differences between very preterm and term-born children.[8] In addition, weaker fine and gross visual-motor skills,[4, 5] reading problems and academic underachievement[9] in very preterm children are associated with visual sensory and perceptive dysfunctions.

The aims of this study were to (1) establish a comprehensive profile of visual functioning by extending routine sensory vision screening with visual perceptive measures, (2) exploring the association between visual sensory and perceptive dysfunctions, and (3) exploring the association between visual abilities and intellectual functioning.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References


A sample of 116 very preterm/very-low-birthweight (VP/VLBW; 57 males, 59 females; median gestational age 30.1wks, range 28.1–31.6; median birthweight 1245g, range 965–1480g) and 73 term children (40 males, 33 females; median gestational age, 40.0wks, range 39.0–40.8; median birthweight 3500g, range 3198–3965) participated in this study. All VP/VLBW children were born in Amsterdam, the Netherlands between 2003 and 2006 and originally participated in a multicentre randomized controlled trial on post-discharge intervention targeting parent–infant interaction and the infant's self-regulation.[10] In this study, we intended to aggregate both groups of the initial trial into one cohort, if no effects of the trial on visual outcome were found. Initially, 176 children born at less than 32 weeks gestational age and/or with birthweight less than 1500g (i.e. VP/VLBW) were included. Exclusion criteria at study entry were severe congenital abnormalities, severe maternal physical or mental illness/problems, not mastering the Dutch language and unavailability of an interpreter, and participation in other trials on post discharge management. Of the 176 infants, 164 were available for follow-up at 5.5 years of age, of whom 136 (77% of the initial sample) agreed to participate. As a result of incomplete data, 20 children were excluded from the analyses, of whom four had serious developmental delay (crucially interfering with task execution), three declined participation, and for 13 children not all data were collected because of time constraints. Of the final VP/VLBW sample (n=116), 63 children participated in the intervention and 53 in the control condition of the initial trial. No evidence was found for differences in terms of perinatal and sociodemographic characteristics between participants (n=116) and non-participants (n=60; p>0.05).

Five-year-old, term-born children were recruited from regular schools attended by the VP/VLBW children (n=30) and additional schools located in the same geographical area (n=43). Term-born children were included if they had a gestational age greater than 37 weeks and birthweight greater than 2500g. Exclusion criteria were severe perinatal complications or illnesses that might interfere with normal brain development, and learning difficulties.

Group characteristics are depicted in Table 1. Perinatal characteristics were taken from the medical records at discharge for VP/VLBW children and reported by parents for the term-born children. Sociodemographic factors were obtained by a parent questionnaire at the 5.5-year assessment. The measure of parental education was derived from the number of years of post-elementary education of both parents and classified as high (either parent >8y), middle (both parents 6–8y), or low (either parent <6y).[11]

Table 1. Group characteristics of participants
CharacteristicsVP/VLBW (n=116)Term (n=73) p a
  1. Data are presented as %, or as median (interquartile range). at-test and chi-square results for continuous and non-continuous data, respectively. bSmall for gestational age defined as <−1SD for Dutch reference norms. cDefined according to Papile et al. dDefined according to de Vries et al. eROP-status was known for 70% children of the VP/VLBW sample. PMA, post-menstrual age; VP/VLBW, very preterm/very-low-birthweight.

Perinatal characteristics
Gestational age, wks30.1 (28.1–31.6)40.0 (39.0–40.8)<0.001
Gestational age ≤28, wks22%  
Birthweight, g1245 (965–1480)3500 (3198–3965)<0.001
Small for gestational ageb24%  
Multiple birth26%  
Antenatal steroid use73%  
APGAR score, at 5min9 (8–10)  
Oxygen at 36 weeks PMA21%  
Postnatal steroid use5%  
Intraventricular hemorrhagec
Grade I–II16%  
Grade III–IV5%  
Periventricular leukomalaciad
Grade I–II10%  
Grade III1%  
Ventricular dilatation4%  
Cranial ultrasound abnormality
Retinopathy of prematuritye
Grade I–II14%  
Grade III–IV5%  
Sociodemographic characteristics
Family status two parents74%  
Parents born in the Netherlands54%  
First language not Dutch12%  
Follow-up characteristics
Corrected age, years5.5 (5.4–5.6)5.6 (5.4–5.8)0.369
Sex, male/female, n57/5940/330.449
Parental education  0.584


Refractive status, oculomotor and visual sensory functions

Refraction was estimated using automated video refraction (Retinomax K-PLUS, RIGHT Medical Products [Virginia Beach VA, USA], non-cycloplegic). Refractive error was defined by prescription glasses, and suspected refractive error at the time or our assessments (i.e. referred for vision assessment at the time of or as a result of our study). Oculomotor assessment included motility (observation of eye pursuit), eye position (cover–uncover test), convergence (moving a small Mickey Mouse toy towards the nose top), nystagmus (observation), and torticollis (observation).

Visual sensory assessment included visual acuity (Lea Single Symbol and Line Chart Tests, at 3m distance), contrast sensitivity (Lea Contrast Cards), binocular visual field (Donder's method), color vision (15-Hue) and stereo vision (Lang II) and was performed while children wore their glasses, if prescribed. Criteria for oculomotor and visual sensory deficits equal the weakest 5% in large cohort and population studies[12-15] (specified in Table 2).

Table 2. Oculomotor and visual sensory functions in very preterm/very-low birthweight children and term-born children
 Visual sensory deficits
 VP/VLBW (n=116)Term (n=73)Statistic p Effect sizeaCriterionVP/VLBWTerm
  1. Results are presented as mean (SD) or %. aPositive values indicate weaker performance of VP/VLBW children. bIncluding manifest deviation only, either convergent, divergent or intermittent, determined with glasses, if prescribed. cFisher's exact test results. dIncluding primary position and end position nystagmus. eIncluding any tilted head position during vision examination. fDepicted as decimal values. arcsec., seconds of arc; deg., degree; VP/VLBW, very preterm/very-low-birthweight.

Oculomotor functions
Eye position, abnormalb10%4%χ²(1)=1.930.165v=0.11Misaligned10%4%
Motility, abnormal2%0 0.523cv=0.08Reduced2%0
Convergence, abnormal5%1% 0.249cv=0.10>7cm5%1%
Nystagmusd2%0 0.523cv=0.08Present2%0
Torticollise3%0 0.285cv=0.10Present3%0
Visual sensory functions
Visual acuity (3m)f
Right eye0.82 (0.17)0.85 (0.18)F1,187=0.690.407d=0.17<0.52%0
Left eye0.81 (0.17)0.87 (0.19)F1,187=4.370.038d=0.33<0.52%0
Binocular0.89 (0.16)0.93 (0.18)F1,187=2.460.119d=0.25<0.500
Binocular, single symbol1.11 (0.20)1.29 (0.33)F1,187=21.17<0.001d=0.70<0.500
Binocular visual field
Superior51 (11)50 (9)F1,187=1.110.292d=−0.10<37 deg.6%1%
Right74 (12)78 (4)F1,187=4.960.027d=0.41<58 deg.4%0
Left76 (9)78 (5)F1,187=3.190.076d=0.26<58 deg.3%0
Inferior60 (12)68 (12)F1,187=17.41<0.001d=0.67<35 deg.3%0
Contrast sensitivity, abnormal01% 0.386cv=0.09>2.5%01%
Stereovision, abnormal12%1%χ²(1)=7.020.008v=0.19>100 arcsec.12%1%
Color vision, atypical34%33%χ²(1)=0.020.883v=0.01Abnormal00
Visual perceptive abilities

Visual coherence sensitivity was measured using two computerized tasks,[16] that required children to indicate a circle on the left or right half of a screen. The circle consisted of either circular oriented lines or coherently rotating dots, during the static and motion tasks respectively, and was presented against a background of randomly organized lines or dots respectively. Coherence threshold values (% coherently displayed lines or dots) were determined after 30 trials using an adaptive staircase procedure and were used as a dependent variable, with lower thresholds indicating better performance.

Visual perceptive abilities were assessed with the Developmental Test of Visual Perception (DTVP-2),[17] encompassing four motor-free subtasks: Position in Space, Figure-Ground, Visual Closure, and Form Constancy. Children had to match a sample figure to the correct answer alternative. For the Position in Space subtask, adjacent items were covered to minimize distraction. Each scale comprised items of increasing difficulty and testing was discontinued after three failures on five consecutive items. Correct items were assigned one point. The sums of points, for each subtask, were used as a dependent variable, with higher scores indicating better performance.

The Face Recognition task[18] required children to indicate if a target photo (presented for 2500ms) was present in a consecutively presented set of four photos, by pressing one of two response buttons (Liberator Ltd, Swinstead, Lincolnshire, UK) with their preferred or non-preferred hand (for trials with the target present or absent, respectively). Photos showed males or females with neutral facial expressions. Children were presented with 40 trials. Mean reaction time and the number of errors were used as dependent variables. For all visual perceptive tasks, deficits were defined as scores worse than the weakest 5% of the term group for those variables that differentiated between VP/VLBW children and term children.


Intellectual abilities were measured using the full scale or a short form[19] of the Wechsler Preschool and Primary Scale of Intelligence for VP/VLBW and term-born children, respectively. For comparability purposes, estimated Verbal IQ (VIQ; Information and Vocabulary) and Performance IQ (PIQ; Block Design and Matrix Reasoning) of both groups were calculated. These estimates correlate by more than 0.9 with VIQs and PIQs obtained from full scale assessment in VP/VLBW children (data not shown).


All available VP/VLBW children were invited for follow-up. Teachers were asked to distribute invitation letters to parents of term children. All tasks were performed in one of two counterbalanced fixed orders (randomly assigned to participating children) and were completed on two separate days, to avoid fatigue.

Tests of oculomotor and visual sensory functions were performed by trained orthoptists. Visual perceptive and cognitive tasks were administered by trained researchers using standardized instructions. The computerized tasks were performed on a Dell Optiplex 2.8 GHz desktop computer with a 17-inch screen. Children were positioned at 40cm from the middle of the screen, to minimize eye movement.

Statistical analyses

All analyses were performed on raw scores, using spss software, version 20.0 (IBM Corp, NY, USA). Before analyzing the VP/VLBW sample collectively, possible effects of the post-discharge intervention on all dependent variables were assessed. Intervention effects and differences between VP/VLBW and term children were analyzed using chi-square, r Fisher's exact tests, or analysis of variance.

Associations between visual sensory and perceptive dysfunctions, and between visual sensory or perceptive dysfunctions and intelligence were assessed using regression analyses. Independent variables entered were group (step 1) and visual sensory functioning (step 2) in predicting visual perceptive functioning, and group (step 1), visual sensory (step 2) and perceptive functioning (step 3) in predicting VIQ and PIQ. Independent variables were preselected to retain sufficient power. Preselection included variables that differentiated between groups (p<0.15) and were associated with the dependent variable (p<0.15). Dependent variables were visual perceptive measures that differed between groups, and VIQ and PIQ. Confidence intervals were bootstrapped.[20] Assumptions of normality, linearity, and homoscedasticity were assessed by means of standardized residuals, and multicollinearity was evaluated using variance inflation factor and tolerance values.[21] Multicollinearity caused by interaction terms was present in two regression models and was solved by centering the independent variables and using these centered variables in calculating the interaction terms.[22]

Because there were 12 twins and two triplets in the VP/VLBW group, analyses were repeated with only one, randomly selected, multiplet member. Alpha was set at 0.01 and significance testing was two-sided. Effect sizes were calculated in terms of Cramer's V for chi-square analysis, Cohen's d for t-tests or Cohen's f2 for regression analysis, and considered small, medium and large (Cramer's V: 0.1, 0.3, and 0.5 respectively, Cohen's d: 0.20, 0.50, and 0.80 respectively, Cohen's f2: 0.02, 0.15, and 0.35 respectively).[23]

The study protocol was approved by the medical ethics committee of the Academic Medical Center, Amsterdam, and all parents signed informed consent.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Sample characteristics

No differences between VP/VLBW intervention (n=63) and VP/VLBW control children (n=53) on any dependent variable were found, with the single exception of a smaller left side of the visual field in the VP/VLBW intervention group (p=0.004), that was not associated with cerebral ultrasound abnormalities and regarded a chance finding. Consequently, all VP/VLBW children were analyzed collectively in subsequent analyses. No evidence was present for differences in age, sex, and parental education between VP/VLBW and term children (p>0.05; Table 1).

Refractive status, oculomotor, and visual sensory functioning

Sixteen (14%) VP/VLBW children and three (4%) term children had prescription glasses during the assessments. An additional 23 (20%) VP/VLBW and seven (10%) term children had to repeat regular vision screening in well-baby clinics or were referred for full eye and refractive examination after the current study, on the basis of suboptimal test results. Table 2 shows the results for oculomotor and visual sensory functioning, indicating no evidence for differences between VP/VLBW children and term-born children in abnormalities in eye position (p=0.165), motility (p=0.523), convergence (p= 0.249), nystagmus (p=0.523), and torticollis (p=0.285).

Line chart visual acuity for the right (p=0.407) and left eye (p=0.038), as well as binocularly (p=0.119) indicated no evidence for differences between VP/VLBW children and term children. Mean binocular single symbol acuity was lower in VP/VLBW children compared with term children (p<0.001). Overall, term children showed higher visual acuity than VP/VLBW children, most prominent using uncrowded charts. The mean inferior binocular visual field was reduced in VP/VLBW children (p<0.001), in contrast to the superior (p=0.292) and left (p=0.076) and right fields (p=0.027). Differences for contrast sensitivity (p=0.386) and color vision (p=0.883) were not found. A higher percentage of VP/VLBW children showed abnormal stereovision compared with term children (p=0.008), with 11 of 14 VP/VLBW children with abnormal stereovision also having abnormalities in eye position, convergence, or refractive errors. Visual sensory deficits affected VP/VLBW children more often (n=32, 28%), than term children (n=7, 10%; p=0.003), and were associated with ROP-status (p=0.012) and severe cranial ultrasound abnormalities (i.e. intraventricular hemorrhage grade III-IV, periventricular leukomalacia grade III, or ventricular dilation; p=0.009).

Visual perceptive abilities

Visual perceptive abilities are depicted in Table 3. Higher mean thresholds for static (d=0.49, p=0.001), but not for motion coherence (p=0.138) were found, indicating lower sensitivity for static visual coherence in VP/VLBW children compared with term children The DTVP-2 subtasks showed weaker performance for VP/VLBW children, as indicated by lower mean Position in Space subtest scores (d=0.42, p=0.006), indicating a selective disadvantage in the ability to distinguish between rotated figures. No evidence was found for group differences in terms of Figure-Ground (p=0.110), Visual Closure (p=0.140) and Form Constancy (p=0.545) subtests performance, as well as for mean reaction time (p=0.124) and accuracy (p=0.699) for Face Recognition. Visual perceptive deficits were found in 20 (17%) VP/VLBW children and, consistent with cut-off criteria, six (8%) term children (p=0.080) and were associated with ROP-status (p=0.006), but not with other perinatal characteristics.

Table 3. Visual perceptive abilities of very preterm/very-low birthweight children and term-born children
Visual abilityVP/VLBW (n=116)Term (n=73)Statistic p Effect sizea
  1. Results are presented as mean (SD). aStandardized mean differences (Cohen's d); positive values indicate weaker performance of VP/VLBW children. bFor VP/VLBW: n=99; for term children n=71. DTVP-2, Developmental Test of Visual Perception 2nd edition; VP/VLBW, very preterm/very-low-birthweight.

Visual coherence
Static22.3 (4.9)19.8 (4.9)F1,187=10.950.0010.49
Motion37.8 (11.9)35.2 (10.7)F1,187=2.220.1380.22
Position in space15.1 (4.5)17.1 (5.2)F1,187=7.620.0060.42
Figure-ground10.4 (2.7)11.1 (3.0)F1,187=2.580.1100.24
Visual closure5.8 (3.3)6.6 (3.8)F1,187=2.200.1400.22
Form constancy10.7 (2.4)10.9 (2.8)F1,187=0.370.5450.09
Face recognitionb
Reaction time (ms)3019 (565)2882 (578)F1,168=2.400.1240.23
Errors (number)12.0 (6.9)11.5 (6.7)F1,168=0.150.6990.02

All analyses for oculomotor, visual sensory, and perceptive functioning were repeated without children suspected of uncorrected refractive error, leaving the results essentially unchanged.


VP/VLBW children had lower mean estimated VIQ than the term-born children: 99.0 (SD 14.8) and 109.8 (SD 15.3) respectively, d=0.71, p<0.001, and lower mean estimated PIQ: 96.6 (SD 15.2) and 106.5 (SD 15.0) respectively, d=0.66, p<0.001.

Association between visual sensory and perceptive dysfunctions

Regression models indicated small-sized associations between visual sensory characteristics (refractive error, left eye, binocular and single symbol acuity, and stereovision), and static coherence sensitivity (ΔR2=0.08, p=0.005, f2=0.006) and Position in Space performance (ΔR2=0.07, p=0.001, f2=0.005). Although the overall models showed significant associations, only group-status and none of the visual sensory characteristics contributed to the variance in visual perceptive functioning. Similarly, at deficit level, seven (35%) of the 20 VP/VLBW children and none of the six term children with perceptive deficits showed accompanying sensory deficits.

Association between visual dysfunctions and intelligence

Table 4 depicts the association between visual functioning on one hand, and VIQ and PIQ on the other hand. Visual sensory characteristics (left eye, binocular and single symbol acuity, inferior and right visual field, and stereovision) accounted for 10% of the variance in VIQ (ΔR2=0.10, p=0.003, f2=0.01), whereas perceptive abilities (static and motion coherence, Position in Space, Figure-Ground, and Visual Closure) contributed an additional 13% of variance explained (ΔR2=0.13, p<0.001, f2 = 0.02). Visual sensory functioning (prescription glasses, left eye, binocular and single symbol acuity, and stereovision) contributed 3% of the variance in PIQ (ΔR2=0.03, p=0.213, f2=0.001), whereas perceptive abilities (static and motion coherence, Position in Space, Figure-Ground, and Visual Closure) accounted for 35% of the variance explained (ΔR2=0.35, p<0.001, f2=0.16).

Table 4. Associations between visual sensory and perceptive functions, and Verbal and Performance IQ in a three-step regression analysis
β (95% CI)β (95% CI)β (95% CI)
Verbal IQ
VP/VLBW−0.30 (−0.44 to −0.16)bVP/VLBW−0.26 (−0.40 to −0.13)bVP/VLBW−0.18 (−0.34 to 0.06)a
  Visual sensory measuresVisual sensory measures
  Acuity left eye−0.20 (−0.43 to 0.05)Acuity left eye−0.30 (−0.53 to −0.04)
  Binocular acuity0.20 (0.03 to 0.44)Binocular acuity0.17 (0.00 to 0.36)
  Single symbol acuity0.20 (0.03 to 0.31)Single symbol acuity0.19 (0.03 to 0.29)
  Visual field, inferior0.02 (−0.11 to 0.16)Visual field, inferior0.11 (−0.00 to 0.24)
  Visual field, right0.06 (−0.11 to 0.17)Visual field, right−0.00 (−0.15 to 0.12)
  Stereovision0.08 (−0.05 to 0.21)Stereovision0.10 (0.05 to 0.24)
  VP/VLBW*stereovision−0.23 (−0.31 to −0.14)aVP/VLBW*stereovision−0.23 (−0.32 to −0.14)a
    Visual perceptive measures
    Static coherence−0.07 (−0.23 to 0.08)
    Moving coherence−0.12 (−0.27 to 0.01)
    Position in space0.09 (−0.05 to 0.25)
    Figure-ground0.21 (0.04 to 0.34)a
    Visual closure0.11 (−0.06 to 0.26)
 R2=0.09b R2=0.19bR2=0.10)a R2=0.32bR2=0.13)b
  1. CI, bootstrapped confidence interval; VP/VLBW, very preterm/very-low-birthweight. ap<0.01. bp<0.001.

Performance IQ
VP/VLBW−0.34 (−0.47 to −0.20)bVP/VLBW−0.30 (−0.46 to −0.15)bVP/VLBW−0.21 (−0.32 to −0.01)a
  Visual sensory measuresVisual sensory measures
  Glasses0.06 (−0.05 to 0.17)Glasses0.01 (−0.13 to 0.13)
  Acuity left eye0.15 (−0.04 to 0.35)Acuity left eye0.02 (−0.13 to 0.18)
  Binocular acuity0.01 (−0.16 to 0.21)Binocular acuity−0.03 (−0.16 to 0.10)
  Single symbol acuity−0.03 (−0.20 to 0.16)Single symbol acuity−0.05 (−0.19 to 0.09)
  Stereovision−0.05 (−0.19 to 0.08)Stereovision−0.04 (−0.17 to 0.08)
    Visual perceptive measures
    Static coherence−0.08 (−0.18 to 0.03)
    Motion coherence−0.14 (−0.26 to −0.01)a
    VP/VLBW*motion coh.0.10 (−0.03 to 0.21)
    Position in space0.28 (0.13 to 0.45)a
    Figure-ground0.22 (0.10 to 0.37)a
    Visual closure0.17 (0.01 to 0.31)
 R2=0.11b R2=0.14bR2=0.03) R2=0.49bR2=0.35)b

All analyses were repeated with only one, randomly selected multiplet member, leaving the results essentially unchanged.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This study compared visual sensory and perceptive functions between VP/VLBW children and term children, and explored associations between sensory and perceptive dysfunctions, and between visual functions and intelligence. We found a higher percentage refractive errors, and visual sensory dysfunctions in VP/VLBW children than in term children, including medium sized effects for visual acuity, stereovision, and inferior visual field, consistent with previous studies.[4, 5] Inferior visual field impairment has been associated with cerebral white matter damage in VP/VLBW children.[24] Visual sensory deficits occurred more often in VP/VLBW children with severe cranial ultrasound abnormalities or ROP, consistent with existing evidence.[1] Visual perceptive dysfunctions in VP/VLBW children were indicated by small to medium effects for Position in Space performance and static coherence sensitivity. Weaker visual-spatial analysis as indicated by the Position in Space subtest agrees with results in a recent meta-analysis.[6] Similar to existing evidence,[25, 26] we found decreased mean static coherence sensitivity in VP/VLBW children, but no evidence of a group difference for motion coherence sensitivity was seen. Weaker motion coherence processing has been found in studies that included smaller samples of older VP/VLBW children,[25, 27, 28] but differences in stimulus configuration may also have caused contradicting findings across studies.[25, 27]

Associations between visual sensory and perceptive dysfunctions were small sized. Only one in three VP/VLBW children with perceptive deficits would have been detected if routine vision screening relies on visual sensory measures, suggesting that routine vision screening should also include visual perceptive tests. Visual sensory and perceptive abilities, each, explained small amounts of variance in VIQ. Interestingly, visual perceptive measures only significantly contributed to the medium sized association with PIQ. In addition to the study[7] that identified visual-spatial abilities among a set of neurocognitive abilities to fully account for IQ differences between VP/VLBW and term children,[7] our results suggest differential effects, favoring a relationship between visual perceptive abilities and PIQ. Specifically, visual perceptive functioning and PIQ both draw on abilities such as discriminating visual reversals and rotations and might share neural networks, including posterior-frontal white matter association tracts and the medial temporal region.[29, 30]

Neural correlates underlying visual dysfunctions in VP/VLBW children are scarcely studied. A specific dorsal stream vulnerability in VP/VLBW children has been suggested, referring to impaired functions of the dorsal visual processing stream (i.e. visual-spatial analysis, motion perception, and unconscious control for visual-motor action).[31] Our results confirm weaker visual-spatial performance, as indicated by the Position in Space test, but we did not demonstrate impaired motion-related perception. In addition, it should be noted that static and motion coherence sensitivity have been shown to rely on dorsal as well as ventral steam functioning.[32]

This study has some limitations. Because of non-cycloplegic video refraction, refractive errors were not detected reliably.[33] Prescribed glasses and current referrals for refractive assessment were used as measures of refractive error instead, and potential adverse effects on visual functioning were evaluated. Besides, reduced visual functions have also been found in VP/VLBW children with adequate refractive correction.[34] Measures of oculomotor functioning, contrast and stereovision may have lacked sensitivity to detect subtle dysfunctions. Reduced sensitivity has possibly led to a low incidence of deficits and, consequently, limited power to detect group differences. Nonetheless, our assessment has been consistent with commonly used tests in clinical (sensory) vision screening of young children and our results indicate that the sensitivity of such tests to detect visual perceptive dysfunctions is very limited. Furthermore, color vision testing based on pseudoisochromatic plates using basic shapes could be more appropriate. The distinction between visual sensory and perceptive measures might be seen as arbitrary, but is consistent with the distinction between ophthalmic and neurocognitive assessments, respectively. Impaired attention functioning might be associated with visual as well as intellectual development, but attention deficits seem unlikely to account for the specific profile of visual deficits that we found.

In conclusion, we found evidence for visual sensory and perceptive dysfunctions in VP/VLBW children compared with term children that occurred independently of each other. In addition, visual perceptive abilities showed a medium sized association with PIQ, in contrast to a small sized association with VIQ. Future studies should investigate factors underlying the associations found, including visual attention dysfunctions, and identify opportunities for intervention.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We are most grateful to Prof. J Atkinson, Prof. O Braddick, and Dr J Wattam-Bell for sharing the static and motion coherence tasks. We thank M Apeldoorn, M Meijer, and S van der Zwet-Slotemaker for conducting the orthoptic assessments. This research was supported by a grant from the Novum Foundation; a non-profit organization providing financial support to (research) projects that improve the quality of life of individuals with visual impairment ( The study sponsor had no involvement study design, data collection and analysis, manuscript preparation, and the decision to submit the paper for publication. The authors have stated that they had no interests that might be perceived as posing a conflict or bias.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    O'Connor AR, Stephenson T, Johnson A, et al. Long-term ophthalmic outcome of low birth weight children with and without retinopathy of prematurity. Pediatrics 2002; 109: 128.
  • 2
    Volpe JJ. Brain injury in premature infants: a complex amalgam of destructive and developmental disturbances. Lancet Neurol 2009; 8: 11024.
  • 3
    Edmond JC, Foroozan R. Cortical visual impairment in children. Curr Opin Ophthalmol 2006; 17: 50912.
  • 4
    Cooke RWI, Foulder-Hughes L, Newsham D, Clarke D. Ophthalmic impairment at 7 years of age in children born very preterm. Arch Dis Child 2004; 89: F24953.
  • 5
    Jongmans M, Mercuri E, Henderson S, deVries L, Sonksen P, Dubowitz L. Visual function of prematurely born children with and without perceptual-motor difficulties. Early Hum Dev 1996; 45: 7382.
  • 6
    Geldof CJA, van Wassenaer AG, de Kieviet JF, Kok JH, Oosterlaan J. Visual perception and visual-motor integration in very preterm and/or very low birth weight children: a meta-analysis. Res Dev Disabil 2012; 33: 72636.
  • 7
    Rose SA, Feldman JF, Jankowski JJ, Van Rossem R. Basic information processing abilities at 11 years account for deficits in iq associated with preterm birth. Intelligence 2011; 39: 198209.
  • 8
    Kerr-Wilson CO, Mackay DF, Smith GCS, Pell JP. Meta-analysis of the association between preterm delivery and intelligence. J Public Health (Oxf) 2012; 34: 20916.
  • 9
    Litt J, Taylor HG, Klein N, Hack M. Learning disabilities in children with very low birthweight. J Learn Disabil 2005; 38: 13041.
  • 10
    Koldewijn K, Wolf M-J, van Wassenaer A, et al. The infant behavioral assessment and intervention program for very low birth weight infants at 6 months corrected age. J Pediatr 2009; 154: 338.e2.
  • 11
    Potharst ES, van Wassenaer AG, Houtzager BA, van Hus JWP, Last BF, Kok JH. High incidence of multi-domain disabilities in very preterm children at five years of age. J Pediatr 2011; 159: 7985.
  • 12
    Robaei D, Rose K, Ojaimi E, Kifley A, Huynh S, Mitchell P. Visual acuity and the causes of visual loss in a population-based sample of 6-year-old Australian children. Ophthalmology 2005; 112: 127582.
  • 13
    Junghans B, Kiely PM, Crewther DP, Crewther SG. Referral rates for a functional vision screening among a large cosmopolitan sample of australian children. Ophthal Physiolog Opt 2002; 22: 1025.
  • 14
    Hargadon DD, Wood J, Twelker JD, Harvey EM, Dobson V. Recognition acuity, grating acuity, contrast sensitivity, and visual fields in 6-year-old children. Arch Ophthalmol 2010; 128: 704.
  • 15
    Aring E, Gronlund MA, Andersson S, Hard AL, Ygge J, Hellstrom A. Strabismus and binocular functions in a sample of swedish children aged 4–15 years. Strabismus 2005; 13: 5561.
  • 16
    Atkinson J, Braddick O. Dorsal stream vulnerability and autistic disorders. Cah Psychol Cogn 2005; 23: 4958.
  • 17
    Hamill DD, Pearson NA, Voress JK. Developmental Test of Visual Perception, 2nd edn. Austin, TX: Pro-Ed, 1993.
  • 18
    De Sonneville LMJ, Verschoor CA, Njiokiktjien C, Op het Veld V, Toorenaar N, Vranken M. Facial identity and facial emotions: speed, accuracy, and processing strategies in children and adults. J Clin Exp Neuropsychol 2002; 24: 20013.
  • 19
    Sattler JM. Assessment of Children: Cognitive Applications. La Mesa, CA: JM Sattler, Publisher Inc, 2001.
  • 20
    Efron B, Tibshirani R. Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy. Stat Sci 1986; 1: 5475.
  • 21
    Tabachnick BG, Fidell LS. Using Multivariate Statistics. Boston: Allyn and Bacon, 2001.
  • 22
    Aiken LS, West SG. Multiple Regression: Testing and Interpreting Interactions. Thousand Oaks, CA: SAGE Publications Inc, 1991.
  • 23
    Cohen J. A power primer. Psychol Bull 1992; 112: 1559.
  • 24
    Jacobson L, Flodmark O, Martin L. Visual field defects in prematurely born patients with white matter damage of immaturity: a multiple-case study. Acta Ophthalmol Scand 2006; 84: 35762.
  • 25
    Taylor NM, Jakobson LS, Maurer D, Lewis TL. Differential vulnerability of global motion, global form, and biological motion processing in full-term and preterm children. Neuropsychologia 2009; 47: 276678.
  • 26
    Atkinson J, Braddick O. Visual and visuocognitive development in children born very prematurely. Prog Brain Res 2007; 164: 12349.
  • 27
    MacKay TL, Jakobson LS, Ellemberg D, Lewis TL, Maurer D, Casiro O. Deficits in the processing of local and global motion in very low birthweight children. Neuropsychologia 2005; 43: 173848.
  • 28
    Guzzetta A, Tinelli F, Del Viva MM, et al. Motion perception in preterm children: role of prematurity and brain damage. NeuroReport 2009; 20: 133943.
  • 29
    Kravitz DJ, Saleem KS, Baker CI, Mishkin M. A new neural framework for visuospatial processing. Nat Rev Neurosci 2011; 12: 21730.
  • 30
    Deary IJ, Penke L, Johnson W. The neuroscience of human intelligence differences. Nat Rev Neurosci 2010; 11: 20111.
  • 31
    Braddick O, Atkinson J, Wattam-Bell J. Normal and anomalous development of visual motion processing: motion coherence and ‘dorsal-stream vulnerability’. Neuropsychologia 2003; 41: 176984.
  • 32
    Braddick OJ, O'Brien JM, Wattam-Bell J, Atkinson J, Turner R. Form and motion coherence activate independent, but not dorsal/ventral segregated, networks in the human brain. Curr Biol 2000; 10: 7314.
  • 33
    Paff T, Oudesluys-Murphy AM, Wolterbeek R, et al. Screening for refractive errors in children: the plusoptix s08 and the retinomax k-plus2 performed by a lay screener compared to cycloplegic retinoscopy. J Am Assoc Pediatr Ophthalmol Strab 2010; 14: 47883.
  • 34
    O'Connor AR, Stephenson TJ, Johnson A, et al. Visual function in low birthweight children. Br J Ophthalmol 2004; 88: 114953.