Abstract
A lab experiment was carried out in order to establish whether pseudoneglect spatial biases could be altered using visuomotor feedback training (VFT) in the form of rod lifting. Previous research shows that those suffering with neglect after right-hemisphere stroke struggle to attend to stimuli on the left side of space. However, this skew can be altered using VFT, allowing those suffering with neglect to attend to the left side of space. Pseudoneglect is the tendency for healthy individuals to consistently bisect lines to the left of true centre. The current study aimed to establish whether this shift in spatial bias could be replicated in healthy populations that show this pseudoneglect. Furthermore, the study aimed to investigate whether those with Autistic traits, show a difference on baseline measures of line bisection. As previous research suggests spatial and processing differences in those with Autism, which may translate within line bisection. Line bisection scores were operationalised as the mean error from the true centre before and after VFT, these errors where analysed and compared using a one-way ANOVA. Results showed there to be no significant change in line bisection error after VFT. Furthermore, a one-sample t-test demonstrated no presence of pseudoneglect within the population at baseline. Autistic traits and line bisection errors were analysed using a two-way Pearson’s correlation, which again demonstrated no significant relationship between these two variables. The results of this study greatly contradict previous research, which strongly suggests that pseudoneglect is consistently visible in healthy populations. Previous research also supports VFT to produce a significant shift in spatial biases within neglect; this would suggest similar effects to be present within the use of VFT for pseudoneglect. However, this was not seen. The lack of significance found for rod-lifting (VFT) and spatial biases, could be attributed to the total lack of pseudoneglect observed at baseline. With this in mind, it may be beneficial to replicate the current study in a population that definitively shows pseudoneglect at baseline, in order to establish whether VFT can induce shifts in spatial awareness, but only when pseudoneglect is present at baseline. Furthermore, it could simply be that such shifts in spatial awareness are a phenomenon only achieved and observed within stroke populations. Finally, prior literature strongly suggests those with Autism to show a lack of pseudoneglect, as well as strong spatial differences in comparison to neurotypicals. However, no support for these theories was found in the current study. It may be necessary to replicate this study using more robust measures of spatial awareness, as differences in spatial awareness within Autism, may not be visible within line bisection.
Keywords: Pseudoneglect, Visuomotor Feedback Training, Autism Spectrum Disorder, Spatial biases, Line Bisection.
Visuomotor Feedback Training on Pseudoneglect in Healthy Populations and those with Autistic Traits.
Autism
Autism Spectrum Disorder (ASD) is classed as a neurodevelopment disorder; neurodevelopmental disorders are defined as those that see their onset within early developmental periods of a person’s life (American Psychiatric Association, 2013). Autism usually presents in infancy, latest at an age of three years old (Hyman, 2001; Lord, Cook, Leventhal & Amaral, 2000). Autism is known as a heterogeneous condition, in which no two people with Autism show the same characteristics (Hyman, 2001). Despite this, symptoms will generally fall into one of two main criteria (American Psychiatric Association, 2013; Hyman, 2001). The first is poor social communication and interactions, with examples of such including failure to initiate and respond to social interactions, a lack of nonverbal communication and deficits in developing, maintaining or understanding relationships (American Psychiatric Association, 2013). The second presented criteria for a diagnosis of ASD is restrictive and repetitive patterns of behaviours, interests or activities (American Psychiatric Association, 2013).
In terms of prevalence, ASD diagnosis is steadily on the increase (Blaxill, 2004; Weintraub, 2011). Steady rises in diagnosis have been observed since the 1970s (Weintraub, 2011), with diagnoses tripling in the United Kingdom (UK) between 1980 and 1990 (Blaxhill, 2004). Similarly, in the United States, diagnoses increased tenfold in the US between 1970 and 1990 (Blaxhill, 2004). In fact, prevalence of Autism in the UK has shown to be higher than in the US (Ratajczak, 2011). However, debate occurs over the actual figures of incidence due to subjectivity in diagnosis criteria used for classification (Ecker et al., 2010; Ratajczak, 2011). Research speculates that increases in the prevalence of ASD may be due to increased education and heightened awareness of Autism, in both parents and developmental specialists (Elsabbagh et al., 2012; Weintraub, 2011; Wing & Potter, 2002).
Aetiology of Autism
When looking at aetiology of Autism, there is no specific known cause (Grinker, 2016). However, Autism has an extremely high heritability (Frith & Happé, 2005). A large proportion of research looking at aetiology suggests gene mutations or deletions to be at the root of autism (Ratajzcak, 2011). However, a wide range of environmental factors have also been linked to the expression of ASD. Inflammation of the brain, due to prenatal exposure to virus and infection and pesticides (Ratajczak, 2011), is discussed as a possible contributing factor in the aetiology of ASD.
Interestingly, newer MRI research has identified increased brain size, in both weight and volume, to be associated with autism (Courchesne & Pierce, 2005). However, further research suggests that this increased volume decreases with time to match the volume of a typically developing cohort (Lange et al., 2015). Despite this, there do seem to be clear biological differences, possibly due to these observed differences in the brains of those with ASD (Hazlett et al., 2012). Further biological contributions, such as increases in parental age, is cited as a possible contributing factor within the aetiology of Autism (Weintraub, 2011).
Processing differences in Autism
While ASD is prominently diagnosed using social symptoms, sensory differences, in the processing of sensory information are also evident (Coulter, 2009; Simmons et al., 2009). Research demonstrates a distinct cognitive profile within those diagnosed with ASD, interestingly, cognitive performance in those with ASD vary from superior to impaired (Coulter, 2009).
Despite these differences within ASD, general patterns and styles emerge in cognitive style. Of particular interest is visuospatial abilities. Visuospatial abilities, defined as the automatic mechanisms used to process visual stimuli (Scott & Schoenberg, 2011), are consistently shown to be different within ASD. Research shows visuospatial abilities in those with Autism to be heightened (Mitchell & Ropar, 2004). Those with ASD tend to show complications with integrating local features of visual stimuli, into global pictures (Chabani & Hommel, 2014).
Compound letter tasks (hierarchal figures) depict a larger letter (global picture), made up of smaller letters (local details). When looking at research ASD populations, there is consistent preference for local details within these figures, alongside poorer ability to integrate these local features into a larger global picture (Scherf, Luna, Kimchi, Menshew & Behrmann, 2008). Shah and Frith (1983) first demonstrated that during embedded figures tests (EFT), in which smaller shapes are submerged within a larger global picture, show, those with ASD were to be superior at identifying local embedded shapes (Happé, 2013; Shah & Frith, 1983). Furthermore, those who score higher on the Autism Quotient (AQ), also show higher accuracy and speed on the EFT (Grinter, Van Beek, Mayberry & Badcock, 2008).
Block design tests (BDTs) require participants to view an image and then reconstruct the shown image with blocks (Wahlstrom et al., 2016). Research consistently shows superior performance during these BDTs from ASD populations (Kana et al., 2013; Shah & Frith, 1993).
Such results have led to a hypothesis known as Weak Central Coherence (WCC) theory, and is theorised to be distinct cognitive style within Autism, in which local elements of stimuli are processed preferentially (Happé, 1999; Shah & Frith, 1993). Frequently, EFT results supporting Shah & Frith (1993) are observed, with WCC being strongly supported in empirical evidence (Happé & Frith, 2006). However, some research does dispute this claim, with research finding similar performances between typically-developing and ASD children (White & Saldaña, 2011).
As seen in BDTs, EFTs and compound figures, there are clear cognitive differences within ASD. These differences are also seen within visual illusion, with ASD populations showing less susceptibility to visual illusions (Simmons et al., 2009).
Neurotypical individuals consistently show leftward attentional biases, however, those with Autism have also been shown to attend less to the left side of centrally presented stimuli (English, Mayberry & Visser, 2015). Similarly, these reduced leftward biases are also visible on greyscale tasks (English, Mayberry & Visser, 2017). This consistent finding of a reduced leftward bias is thought to be due to a lateralisation of attention (English, Mayberry & Visser, 2017). Interestingly, there is another population that displays this lateralisation of spatial attention; Hemispatial neglect patients with right hemisphere damage (English, Mayberry & Visser, 2017).
Neglect
Hemispatial neglect is a neurological condition in which those affected cannot attend to the contralesional side of space, in terms of perception orientation and movement to the left side (c; Rossit et al., 2017; Serino, Barbiani, Rinaldesi & Làvadas, 2009). It is an extremely common condition after stroke, in particular, after right-hemisphere stroke (Husain, 2008). In fact, Buxbaum et al. (2004), found neglect to be present in 48% of a 166 sample of patients, after right-hemisphere stroke and further research estimates hemispatial neglect to be present in between 50-70% of those who suffer a right-hemisphere stroke (Goedert, Zhang & Barrett, 2015).
Despite not being able to attend to one side of the visual field, normally the left (Goedert, Zhang & Barrett, 2015), interventions have been created that reduce such harsh effects of neglect. One of these interventions is prism adaptation (Goedert, Zhang & Barrett, 2015). Prism adaptation (PA) uses goggles that induce a shift in the visual field to the right (Serino et al., 2009). Research shows that repeated visuomotor tasks, such as pointing and movements, can lead to compensation of the previously neglected left-side of space (Serino et al., 2009). Moreover, prism adaptation has shown to improve biases in representational and mental imagery (Rode, Rossetti & Boisson, 2001), as well as improve rightward skewed posture and balance (Hideki, Yuriko, Ayaka, Toshiaki & Takashi, 2008). However, sample sizes of neglect patients are often extremely small, alongside variations in results of PA’s effectiveness (Fortis, Goedert & Barrett, 2011).
A further intervention used is visuomotor feedback training. Using rods within repeated visuomotor feedback training (VFT), patients must grasp to lift rods at the centre (Rossit et al., 2017). Improvement in this task has observed after three days, and after a one month follow up; showing long-term improvement in hemispatial neglect symptoms (Harvey, Hood, North & Robertson, 2003). Furthermore, Rossitt et al. (2017), again found VFT to produce both short and long-term improvements on line bisection, as well as reducing spatial biases within cancellation tasks.
Pseudoneglect
As seen, neglect is a serious neurological bias, normally affecting those who suffer a stroke (Buxbaum et al., 2004). However, a form of spatial bias is also observable within neurologically healthy populations. Pseudoneglect is a phenomenon in which neurologically healthy populations display a consistent leftward attentional bias, in terms of both mental and physical space (English, Mayberry & Visser, 2017; Schmitz & Peigneux, 2011). Meta-analysis clearly demonstrates that within line bisection tasks, neurologically healthy participants consistently incorrectly bisect lines to the left of the verdical centre (Jewel & McCourt, 2000). It is theorised that this leftward bias may be due to right hemisphere dominance and processing (Benwell & Thut, 2013; Schmitz & Peigneux, 2011).
Interestingly, those diagnoses with ASD tend not display these attentional leftward biases. In fact, it has been shown that those with ASD actually attend less to the left side of centrally presented visual stimuli (English et al., 2015). In a comparison of those with ASD and controls, a negative correlation emerged between social skills score on the AQ and leftward biases in a greyscales task (English et al., 2015). From this, a clear difference in processing and spatial attention is visible within those with autism. It may be, that this reduced leftward bias is due to the superior visuospatial abilities within ASD.
Therefore, as previous research shows, rod lifting VFT can impact upon line bisection task accuracy, in those who suffer from hemispatial neglect. With this in mind, it will be interesting to see whether VFT can induce a spatial bias in healthy individuals. Moreover, it seems as though those with ASD do not display attentional leftward biases of neurotypicals, therefore, VFT may impact upon line bisection tasks in an opposite manner to those with spatial neglect. Not only this, but those with ASD may display a completely different bias on the baseline measure in comparison to neurotypicals. With this in mind, the primary hypothesis for this experiment will be; weighted rod lifting (VFT) will affect line bisection in healthy populations. The secondary hypothesis for this experiment will be; there will be a difference in baseline measure of pseudoneglect in those with autistic traits compared to controls.
Methods
Design
An experimental design was employed in order to explore the effects of visuomotor feedback training task (weighted rod-lifting) on line bisection tasks, as well as the effects of autistic traits on baseline line bisection. The study was approved by the UEA ethics committee (Appendix A). This study was part of a group research project, meaning many variables were collected. However, this research will focus only on the effects of visuomotor feedback training (weighted rod lifting) on line bisection error. Alongside this, the effect of Autistic traits on baseline line bisection error.
Within this study, the first independent variable was the rod lifting condition that participants were assigned to. This independent variable had three levels; rightward weighted rod, leftward weighted rod and a control with a centre weight. The second independent variable within this research was Autism, operationalised as a score using the Autism Spectrum Quotient (Baron-Cohen et al., 2001) (Appendix B).
The dependent variable within this study was the accuracy on line bisection tasks before and after rod lifting. This variable was operationalised as the mean error from the true centre of the line, across the line bisection trials. A positive error indicates a rightwards bias and a negative error indicates leftward bias.
Participants
Participants were recruited through the use of posters around the campus of the University of East Anglia (Appendix C), the use of these posters alongside word of mouth, allowed the recruitment of 61 participants. Participants were gathered on a volunteer opportunity basis, this therefore means that a large proportion of the sample are students studying at the University of East Anglia. A total of 61 participants completed the survey and laboratory study. However, three participants were excluded due to incorrect completion of distractor questions, a further three were excluded due to being left-handed. The final sample was 55. The age range of participants was 19 to 50, with a mean age of 21.75 (SD= 4.49). Gender within the sample constituted 20 males and 35 females. All remaining participants were right handed, with normal to corrected vision.
Materials
Part A of the current study took the form of an online questionnaire, using a survey design website, Qualtrics. The materials within, consisted of questionnaires and two short exercises. The first online materials given were the information and consent forms (Appendix D and E, respectively), followed by a laptop questionnaire (Appendix F).
Demographics questionnaire recorded variables such as age, gender and race (Appendix G). An adapted version of the Edinburgh Handedness Inventory (Oldfield, 1971), was also used in order to identify handedness. This measure calculates an individual’s handedness based on the hand used for a range of actions, such as writing, opening a box lid, and using a spoon (Appendix H).
Mental rotation tasks, as created by Vandenburg & Kuse (1978) (Appendix I), were a further material used within the online study. In total, there were 25 mental rotation tasks, including an initial example task. These tasks measure how accurate one’s spatial visualisation is (Vandenburg & Kuse, 1978). An example of mental rotation question, can be seen in appendix I.
The Autism Spectrum Quotient (Baren-Cohen et al., 2001) measures symptoms of ASD based on a 50 item self-report. The individual is required to respond to statements such as ‘I prefer to do things with others rather than on my own’, using a four-item scale ranging from definitely agree, to definitely disagree (Appendix B).
The adult ADHD self-report scale (Adler, Kessler, Spence & World Health Organisation, n. d.) (Appendix J), measures symptomatology of ADHD. Individuals are asked to respond to questions in relation to how they have felt over the past six months, for example, ‘How often do you have difficulty getting things in order when you have to do a task that requires organisation?’. This question is responded to using a five-point scale ranging from never to very often.
The final task material used in the online study were 15 line bisection tasks, lines presented either in the centre, left or right of the screen (see Appendix K). Lines were 164.10mm long, white, vertical and presented on a black background. An online debrief form was also distributed (Appendix L).
Part B of the study involved further information and consent forms (Appendix M and N, respectively). Line bisection tasks using E-prime software were also used (Appendix K). A 50cm metal rod was used for the lifting task, within this rod was a partially concealed weight, which could be moved to 21 positions on the rod (+10 to -10) (Appendix O). Materials used within the rod lifting task, included a felt mat (122x50cm), which contained starting and moving position markers for the rod (Appendix P). Line bisection tasks were administered using a touch screen computer (800x500mm), with an adjustable chin rest placed in front of the touch screen (Appendix Q). Lines for this line bisection tasks were 185.93mm long. Debrief forms for part B of the study were also used (Appendix R).
Procedure
Part A (online study):
Participants were able to sign up for part one of the study by emailing the researchers, using the email provided on recruitment posters (Appendix C). Once signing up for the study, participants were sent an anonymous link for the Qualtrics survey. The first part of this survey began with an information sheet and consent form, this showed a box which participants were required to check in order to signal the understanding of the information sheet and agreement to participation. Once consenting, participants were asked to complete administrative questionnaires including a demographics questionnaire (Appendix G), a laptop questionnaire (Appendix F). Participants were then asked to create a unique participant code using the first letter of their mother’s maiden name, the last letter of their surname, and the two-digit date of their birth.
Once these tasks had been completed, participants were asked to complete a set of 25 mental rotation tasks, including an example at the beginning (Appendix I). Following completion of the completion of these tasks, participants were asked to complete the Autism Spectrum Quotient (Baron-Cohen et al., 2001) (Appendix B), followed by the adult ADHD self-report scale (ASRS-v1.1) (Adler, Kessler, Spencer & World Health Organisation, n.d.) (Appendix J). There was no set time limit for the online tasks or questionnaires.
Participants were then asked to complete 15 trials of line bisection. Lines were placed either left, right, or in the middle with five lines per category (See example in Appendix K). Mental rotation tasks and line bisection tasks were randomised in order to combat order effects. After completion, participants were given debrief forms to clarify the study purpose, and whom to contact for further information (Appendix L). Participant codes were indicated on debrief forms for reference in further correspondence if participants wished to withdraw or receive feedback. The total time for part A of the study was approximately 15 minutes. Following the completion of part A of this study, participants were encouraged to sign up for a timeslot for part B (the laboratory study).
Part B (laboratory study):
Participants for the experimental lab portion of the experiment were again recruited through posters and word of mouth. Upon arrival at the lab, participants were seated facing a touch screen and asked to read an information sheet (Appendix M), and sign a consent form (Appendix N). Following this, participants were asked to recall their unique participant code created during part A of the study, in order to allow cross-referencing of data from both parts.
Participants were asked to place their chin in a fixed chin rest, with the left hand in their lap and the right hand on the space bar as a starting and resetting point for each line bisection trial. Participants were asked to complete 15 trials of line bisection on the touch screen (see Appendix K). These line bisection trials were run using E-prime, each line was. Each trial was followed by a mask screen, in order to prevent visual feedback of the line within the eye; therefore, preventing an afterimage that could affect the next trial. After completion of the line bisection tasks, participants were asked to close their eyes, and the experimenter moved the participant over to the rod lifting table (Appendix P).
Prior to arrival, participants were randomly divided into one of three conditions; A control with a centre-weight in the rod, a right-weighted rod group, or a left-weighted rod group (Appendix S). Each participant also had a pre-randomised list of 30 rod lifting positions (Appendix T). This meant that the rod lifting task could be set up prior to arrival and prevented participants from detecting the weight within the rod. Alongside this, the rod was always placed with the weight facedown, so it would not be detected by participants.
Participants were seated with their midline in the centre of a table, the midline of the table was indicated on the surface in order to ensure all participants were seated in the same place. On this table was a felt mat with two starting positions and two lifting positions (Appendix P). The rod had already been placed on the starting position as per the random allocation sheet (Appendix S). Participants were asked to open their eyes and given instructions to lift the rod using a pincer grip, (thumb and index finger) (Rossit et al., 2017), where they believe the rod would balance upon lifting. Participants were told to place the rod back down on the mat and repeat as many times as necessary until happy the rod was balanced. Participants were asked to notify the experimenter upon belief that the rod was balanced. Following this, the rod was moved to the next position and participants were asked to carry out the same task. In total, there were 30 rod lifting trials, using three positions on the felt mat (Appendix P); moving positions one and two and the starting position as moving position three.
Following rod-lifting trials, participants were again asked to close their eyes. Experimenters moved participants back over to the touch screen, where following an identical format, participants once again completed 15 line bisection trials (examples of which can be found in Appendix K). After completion of the second session of line bisection tasks, participants were verbally debriefed and also given a physical debrief (Appendix R). Participants were asked if they had any questions, and were told to email the experimenter should any questions or queries arise. They were also notified of the week notice period for data withdrawal. Part B of the study took 30 minutes to complete.
Results
Hypothesis 1
H1: Weighted rod lifting (VFT) will affect line bisection in healthy populations.
Calculations for Zskew and Zkurtosis (table 1), displayed that scores within the leftwards weighted group were not within the range of normality for the Qualtrics line bisection task (Appendix U). Due to this, Zscores were created for this group, and two outliers were identified within the online line bisection scores and removed. After the removal of these outliers, all Zskew and Zkurtosis scores fell within the ranges of normality (Appendix V).
Descriptive statistics in table 1 shows the average error (mm) for baseline line bisection task (prior to rod lifting), for each experimental condition. These descriptives show all three experimental conditions to display positive error on line bisection tasks. The highest positive error is seen in the left weighted group with a mean error of 2.6 (SD = 2.42), this is extremely similar to the error displayed by the right weighted condition (M = 2.54, SD = 3.02). Mean error for the no weighted rod (neutral condition), shows to be slightly lower at 1.19 (SD = 3.11).
A one sample t-test was used in order to analyse whether these mean error scores were significantly different from the true centre (0). Results showed there to be no significant difference in error from true centre at baseline, t(52) = 1.56 , p = .124.
In order to investigate whether there was a significant difference in errors before and after rod lifting, a parametric repeated measures one-way Analysis of Variance (ANOVA) statistical test was appropriate for analysis.
Table 1
Descriptive statistics for line bisection.
Condition Mean (mm) SD
Line bisection error, pre rod-lifting Left 2.6 2.42
Neutral 1.19 3.11
Right 2.54 3.02
Line bisection error post rod lifting Left 0.19 2.93
Neutral -2.11 3.17
Right 3.08 3.18
Note: N=51.
Table 1. A table to show mean error (mm) across the three conditions, both pre and post rod lifting.
Descriptive statistics in table 1 display that the mean error (mm) for line bisection tasks before rod lifting (baseline), are extremely similar across all three conditions. Descriptive statistics in figure 1 also show considerable changes in error for each group following the weighted rod lifting task. The leftwards weighted group sees a decrease rightwards mean error (mm), from 2.6mm (SD= 2.42), to 0.19mm (SD=2.93). The neutral (no weighted) condition, displays a reverse from rightwards error (1.19mm, SD= 3.11), to a leftwards error of -2.11mm (SD= 3.17). Finally, the rightwards weighted rod-lifting conditions shows an increase in rightwards error from 2.54mm (SD= 3.02), to 3.08mm (SD=3.18).
Changes across this mean error before and after weighted rod lifting for each condition, can be seen in figure 1, 2, and 3, along with error bars displaying the standard error.
Figure 1. A graph to show differences in error before and after visuomotor feedback training for the leftwards weighted rod-lifting group.
Figure 2. A graph to show differences in error before and after visuomotor feedback training for the no weighted (neutral) rod-lifting group.
Figure 3. A graph to show differences in error before and after visuomotor feedback training for the rightwards weighted rod lifting group.
A Levene’s test indicated that the assumption of equality of variances between groups had not been violated for conditions in both the pre line bisection (p = .380), or the conditions for the post line bisection (p = .735), and statistical analysis could continue. A one-way within groups ANOVA showed no significant effect of weighted rod-lifting conditions on error (mm) in line bisection tasks, F(2, 50) = .159, p = .854, ɳp^2 = .006.
Hypothesis 2
H2: There will be a difference in baseline measure of pseudoneglect in those with autistic traits compared to controls.
Looking at descriptive statistics, there does seem to be a large variation in autistic traits of participants, with scores ranging from 2 to 33. The mean ASD score of participants was 16.82 (SD = 7.58), the high standard deviation again indicates a lot of variance in ASD scores of participants.
In terms of line bisection error, participants error scores range from -21.78 to 28.60, with a mean of 2.49 (SD= 11.62). Again, such large ranges in score, along with a high standard deviation suggest much variance in the data.
Figure 4 shows these two variables plotted against each other on a scatterplot. From the graph, there seems to be no emerging pattern of correlation within the data, with data points seemingly randomly distributed across the graph.
Figure 4. Scatterplot to demonstrate the relationship between autistic traits and error (mm) on baseline line bisection.
A two-way Pearson’s correlation, showed no interaction between Autistic traits and error (mm) on the baseline (pre) measure of line bisection; r= -.065, p = .651.
Discussion
The results of the one sample t-test revealed no significant spatial bias at baseline. This result is not in line with previous research, which consistently displays neurotypicals to have an unconscious leftwards bias, known as pseudoneglect (English et al., 2007; Schmitz & Peigneux, 2010). Pseudoneglect has also been supported through meta-analysis (Jewel & McCourt, 2010). From this, the results of the current study are not supported within wider literature.
Past research has also suggested that line bisection tasks do not hold very high sensitivity in terms of measuring and identify neglect (i.e. pseudoneglect and spatial biases) (Azouri et al., 2017). Moreover, it could be possible that pseudoneglect was not visible due to low sensitivity of line bisection tasks. Future research may benefit from a battery of tests investigating spatial biases, in order to more accurately identify neglect and pseudoneglect. However, it must also be mentioned that previous research particularly highlights pen and paper measures of line bisection as having lowest sensitivity. Therefore, it would be an advantage of the current study, that line bisection tasks were digital. It is still important, however, to consider the possibility that this measure may not have best identified biases.
In terms of the effects of rod lifting (VFT) on line bisection, a one-way ANOVA displayed no significant difference in line bisection error or direction after VFT. However, there were definite changes in error within each condition. Despite being non-significant, graphs show the leftwards weighted condition to reduce in error and the rightwards weighted group to increase in error. Interestingly, the neutral weighted (control) group a reverse from rightwards error prior to rod-lifting to leftwards error post rod-lifting.
It is possible that the hypothesis 1 was non-significant due to the absence of pseudoneglect during baseline measure. Previous research shows VFT to be effective at shifting leftwards spatial bias in those with hemispatial neglect (Harvey et al., 2003; Rossit et al., 2017). It is possible that effects of VFT were present in previous research due to an established baseline of pseudoneglect, not seen within the current study. Furthermore, it may be useful, as previously mentioned, to replicate this experiment using a battery of spatial bias measures in order to more accurately identify baseline bias.
Furthermore, it may be said that VFT produces a rare phenomenon within right-hemisphere stroke patients that is not replicable within healthy populations. However, it must also be considered that the results of this experiment may be due to random error, especially considering the inconsistency compared to wider literature (English et al., 2007; Rossit et al., 2017). Therefore, it would be beneficial for this study to be replicated in order to investigate whether the results of this experiment were random error, or whether these VFT shifts in spatial bias are a niche phenomenon within hemispatial neglect patients (Rossit et al., 2017). It is important to replicate this study, as more reliable results would provide important insight into inner workings of spatial awareness within different populations. Furthermore, insight into spatial awareness of healthy individuals and those with neglect, may allow advances of treatments and interventions administered to those suffering with neglect.
ASD did not significantly show relate to line bisection error. However, there does seem to be a negative relationship between Autistic traits and line bisection error. This would suggest that higher autistic traits are related to lower error on line bisection scores. This, despite not being significant, falls in line with previous research that suggests those with ASD do not present pseudoneglect on bisection tasks. Furthermore, previous research suggests that line bisection is not the most sensitive measure of spatial bias and awareness (Azouri et al., 2017). Therefore, it may be necessary to investigate this hypothesis using more robust measures of spatial awareness, and several of them, in order to identify spatial biases in those with Autistic traits. It is important to establish these spatial differences, in order to further knowledge of the extremely varying cognitive profile within Autism (Coulter, 2009). Research also shows strong visuospatial abilities in those with Autism (Mitchell & Ropar, 2004), further investigation of spatial biases within Autism would further advance knowledge of these superior visuospatial abilities, and the circumstances under which they thrive.
Another point to consider is that line bisection, as previously discussed, is not a strong or sensitive measure of spatial biases (Azouri et al., 2017). Interestingly, methods within the literature investigating spatial awareness and biases in both neglect and Autism, tends to employ several measures that investigate spatial awareness and bias (English et al., 2017; Rossit et al., 2017). This may be due to low sensitivity of line bisection, and therefore future research should use several measures of spatial awareness in order to create a comprehensive profile of spatial awareness.