J Manipulative Physiol Ther. 2018 (Feb); 41 (2): 81–91 ~ FULL TEXT
Bassem Farid, BHSc (Hons), Paul Yielder, PhD, Michael Holmes, PhD, Heidi Haavik, PhD, Bernadette A. Murphy, DC, PhD
University of Ontario Institute of Technology,
Oshawa, Ontario, Canada.
OBJECTIVE: The purpose of this study was to test whether people with subclinical neck pain (SCNP) had altered visual, auditory, and multisensory response times, and whether these findings were consistent over time.
METHODS: Twenty-five volunteers (12 SCNP and 13 asymptomatic controls) were recruited from a Canadian university student population. A 2-alternative forced-choice discrimination task with multisensory redundancy was used to measure response times to the presentation of visual (color filled circles), auditory (verbalization of the color words, eg, red or blue), and multisensory (simultaneous audiovisual) stimuli at baseline and 4 weeks later.
RESULTS: The SCNP group was slower at both visual and multisensory tasks (P = .046, P = .020, respectively), with no change over 4 weeks. Auditory response times improved slightly but significantly after 4 weeks (P = .050) with no group difference.
CONCLUSIONS: This is the first study to report that people with subclinical neck pain (SCNP) have slower visual and multisensory response times than asymptomatic individuals. These differences persist over 4 weeks, suggesting that the multisensory technique is reliable and that these differences in the SCNP group do not improve on their own in the absence of treatment.
Key Indexing Terms: Reaction Time, Neck Pain, Superior Colliculi, Choice Behavior
From the Full-Text Article:
Approximately 30% to 50% of people experience neck pain every year, which places a significant burden on the health care system.  A specific category of neck pain is subclinical neck pain (SCNP), which refers to lower-grade neck dysfunction in which individuals have recurrent flare-ups of neck pain but have not yet sought regular treatment. [2–4]
Altered afferent input through repetition and overuse changes the way that sensory information is processed by causing plastic changes in the central nervous system. [5–7] Several studies have provided evidence for altered proprioceptive and neuromuscular function in individuals with SCNP. [8–11] Further research has suggested that the altered afferent input arising from SCNP leads to altered sensorimotor integration (SMI), altered motor output, and impaired motor control. [5, 12–14] It has been hypothesized that other sensory modalities such as visual and auditory inputs may also be processed differently in SCNP causing altered multisensory integration. 
When visual input is less clear (ie, “noisier”), less emphasis is placed on visual information and accuracy is decreased.  Adding haptic stimuli leads to improved accuracy and discrimination when visual input is less reliable.  However, when a sense is less reliable or provides contradictory information, combination of stimuli may not always enhance accuracy. Hogendoorn et al  reported that when there was contradictory input between proprioception of hand location and a visual afterimage, the visual afterimage of the hand disappeared. This suggests that when multisensory input is contradictory, the central nervous system may try to arrange sensory information in relations to other sensory modalities, or even adapt motor output to muscles (eg, less contraction to more accurately reach the target).
In a study with mismatches between proprioception and vision, participants were asked to point toward a target with their unseen hand.17 The relationship between proprioception and vision was manipulated by imposing erroneous visual feedback about the target’s location.  With the distorted visual feedback, participants corrected their responses toward the adjusted visual image up to 60% of the distance of the incorrect visual feedback. When a sensory modality is less reliable or provides contradictory information, less weight is placed on the unreliable sense. [18, 19] Proprioception is impaired in SCNP. [11, 20] However, it is not clear exactly how proprioceptive mismatch affects integration of other senses. It is possible that SCNP may cause greater reliance on senses other than proprioception; however, it is also possible that the less reliable proprioceptive input in SCNP can negatively influence the processing of other sensory stimuli.
Previous research has indicated that multisensory integration in elderly populations is enhanced, as measured using a 2–alternative, forced-choice discrimination task involving measurement of response times.  Another study reported that the elderly have both altered SMI and altered multisensory integration.  Tasks such as temporal order judgements may not be effective at measuring multisensory integration in naïve participants because there is an experience component that “increases” intersensory synchrony ability.  Areas in the thalamus integrate sensorimotor and multisensory processes.  The superior colliculus is 1 example of a multisensory site that has access to efferent projections of premotor and motor areas of the spinal cord and brainstem involved in superior colliculus–mediated attentive and orientation behaviors,  some that involve the saccadic movements of the eye, and neck movements, or “gaze shifts.” In addition to 2 sensory modalities terminating in a target area, a cross-modal projection of 1 sensory modality can also converge onto a different sensory modality site. 
Hairston et al  reported evidence for inverse effectiveness, where asymptomatic participants with artificially degraded vision (induced myopia) had their localization skills enhanced for audiovisual conditions, whereas in normal conditions the localization for multisensory conditions was similar to unisensory conditions. These papers suggest that a similar effect could occur in populations with altered sensory input from the neck, where that multisensory integration may be enhanced, as a result of ongoing alterations in sensory feedback from the neck joints and muscles that lead to degraded somatosensory input, similar to the degraded vision.
Numerous methods can be used to measure multisensory integration. [21, 23, 25, 28] A popular task that uses behavioral enhancements from multisensory integration is a 2–alternative forced-choice discrimination task with semantically congruent, redundant, multisensory stimuli. [21, 28] In some studies researchers accounted for the redundancy gain, a quickening of response times that occurs with redundant stimuli, by using the race model.  When measuring multisensory integration, temporal and spatial factors of the equipment used to present the stimuli must be carefully considered when designing a task, because both timing and location of stimuli may reduce or deplete multisensory activity. [30, 31] Unfortunately, many studies fail to state or consider in detail the equipment being used in multisensory tasks, which can hinder interpretation of results, [21, 23, 28] a limitation that the present study tries to address.
The purpose of this study was to determine if there were differences in multisensory integration and unisensory and multisensory response times in individuals with SCNP compared with asymptomatic participants, and if the results were consistent over a 4–week interval in the absence of treatment for the SCNP group. It was hypothesized that individuals with SCNP would possess slower response times for both unisensory and multisensory conditions because of the ongoing effects of unreliable proprioceptive feedback from the neck.
The results described are the first to reveal that there is a significant difference in response times for visual and multisensory stimuli in a 2–alternative forced-choice discrimination task between adults with SCNP and asymptomatic controls. These differences are not simply because of slower movement times in general, because previous studies of people with SCNP reported no difference in simple response times in this population but found that the SCNP group had slower response times in more complex tasks (mental rotation). 
Results from this study’s SCNP group are similar to the slow unisensory response times reported in the elderly population in the study by Laurienti et al  Though identifying similar group differences between response times, the actual response time values did not match between the studies. All of our mean values and standard deviation for response times for each stimulus condition and were lower than those of Laurienti et al  possibly because we used equipment and drivers that produced a lower latency and variance than those of Laurienti et al. This is supported by the fact that our mean response times for visual conditions for the asymptomatic young adults (396 ± 35 ms) was lower than their young adult controls for the multisensory condition (485 ± 93 ms). We did not remove responses less than 250 ms; however, the large differences in response times between the 2 studies (eg, visual: 396 ± 35 ms vs 538 ± 117 ms) cannot solely be accounted for by this removal because it would only account for a minor difference (eg, <10 ms for visual).
Our standard deviations were almost one-third of the values reported by Laurienti et al  coupled with our faster response times, suggesting that our stimulus presentations were stronger because of a combination of smaller delays and lower variability in timing of stimulus presentation and recording parameters of the equipment used to collect the data. The multisensory gain was not as great as expected because the difference between the visual and multisensory stimulus was extremely small for the control group and SCNP group and there was no significant difference in the multisensory gain between the 2 groups for both the standard t test and CDF analysis. This is dissimilar to some earlier work, [21, 28] which reported a large difference between their visual and multisensory stimuli for all groups. Researchers have pointed to the importance of spatial and temporal factors in the stimulus presentation required to create multisensory integration, [30, 31] which may not have been effective enough in this experiment. However, Meredith and Stein  have reported multisensory integration when there is a delay of up to 100 ms between the 2 multisensory stimuli. The multisensory response times were similar to what the race model had predicted as simply being a redundancy gain. The race model predicts that the quickest response to combined audiovisual stimuli is a “redundant signal effect” that is explained by a summation of the probability of either stimulus being presented. Times faster than this are considered to represent true multisensory integration, rather than just interaction. 
However, Juan et al  recently conducted a study in both rhesus macaque monkeys and human participants. They used a large collection of both unimodal and bimodal audiovisual stimuli with semantic specificities that were presented at different saliencies (weightings). The authors reproduced the redundant signal effect; however, there were no systematic violations of the race model. The authors concluded that the redundancy effect at a neural level depends on how complex the stimuli are, with different consequences for unisensory vs multisensory brain areas.  This suggests that the reason that we did not identify enhanced multisensory gain may have been because our visual and auditory stimuli were not sufficiently complex for multisensory gain to provide any advantage, because our visual response times were already so fast relative to other studies as a result of careful attention to technical factors to minimize latencies of stimulus presentations. 
Most of the time bins revealed that the multisensory gain identified was solely based on the redundancy effect, because most bins did not violate the race model. However, the CDF from the SCNP group did identify multisensory integration beyond what was predicted by the race model between 290 and 320 ms, possibly reflecting increased gain in the SCNP group. However, a significant deviation below the race model prediction (eg, decreased multisensory gain) occurred between 430 and 480 ms, making it hard to draw definitive conclusions about differences in multisensory gain between the 2 groups. Additionally, the longer response times reported by Laurienti et al  suggest that the equipment they used may have had longer lag times in stimulus presentation and recording, which would have effectively acted as “noise,” weakening the unisensory stimulus strength and causing increased reliance on multisensory integration. Noise would have the effect of obscuring the strength of the stimulus signal, paradoxically enhancing multisensory gain, which is strongest when the unisensory stimuli are weaker. 
Individuals with SCNP have altered SMI and altered motor responses.  Treatment of SCNP can improve somatosensory processing as well as elbow joint position sense.  Areas in the thalamus have been reported to be sites of interaction between multisensory integration and SMI.  The superior colliculus is a multisensory area with access to efferent projections of premotor and motor areas of the spinal cord and the brainstem involved in superior colliculus–mediated attentive and orientation behaviors.  A recent study  reported that seniors have altered sensorimotor feedback that may increase their risk of falling. Chiropractic treatment lead to improved sensorimotor and multisensory integration, reducing their risk of falls and suggesting a connection between altered sensorimotor and multisensory integration.
A recent study  reported that changing visual feedback altered the amount of pain-free neck rotation in a group of patients with chronic neck pain. This indicates an increased reliance on visual input in this group, possibly because their internal body schema or body map is not accurately calibrated, leading to altered integration of sensory input. SCNP participants also have worse mental rotation abilities, which likely reflects an altered egocentric frame of reference as a result of an altered body schema.  If patients with neck pain have an increased reliance on vision and their visual and multisensory processing response times are impaired, as indicated in the present study, it is problematic. Our study identified worse visual and multisensory reaction times in SCNP vs asymptomatic young adults, which was not compensated for by increased multisensory gain between the 2 groups, for both the standard t test and CDF analysis. The lack of change over 4 weeks suggests that left untreated, even subclinical neck pain can affect multisensory processing.
Another possibility for the differences in our results compared with previous work such as animal studies may relate to differences in upper level cortical processing. Felines possess efferent projections from the superior colliculus that were enhanced with multisensory input in behaviors involved with attention and orientation.25 Our study suggests that additional factors, such as changes in projections from neck afferents, may interfere with upper-level cortical processing, affecting the initially quick response times of multisensory neurons. As mentioned in the introduction, the superior colliculus integrates sensory inputs that involve movement and orientation behaviors (eg, eye saccade, neck movement toward stimulus).  The altered sensory feedback from the neck in the SCNP group would have projected with other modalities onto these bimodal neurons, or through cross-modal projections onto other sensory sites, and affected multisensory processing.
Our initial hypothesis was based on the concept that neck pain, like aging, would create a noisier system and longer unisensory response times, which we did find in the present study. However, we presumed that like the study by Laurienti et al  the multisensory neurons would have inverse effectiveness and greater gain to compensate. However, it is important to consider that neck pain is different from aging in that some neck afferents project directly to these same multisensory neurons, possibly altering their capacity to increase their gain to compensate for the slower unisensory processing. Previous work reported that spinal manipulation improved SMI in participants with SCNP.  It would be interesting to see whether spinal manipulation is able to improve response times of other unisensory and multisensory modalities.
This study found that adults with SCNP have slower response times for visual and multisensory stimuli than asymptomatic controls, but differences in multisensory integration were inconclusive. The auditory component improved over time, suggesting a need to perform multiple baseline tests to ensure that baseline performance has stabilized before including auditory stimuli in a longitudinal study. Although great care was taken to determine the technical specifications of all equipment, it is possible that technical factors such as screen refresh rates could have introduced delays in some of the response times. That being said, this would have been a random effect for both groups and thus would be unlikely to have affected the final results.
This is the first study to report that people with SCNP possess slower visual and multisensory response times. These differences persist over 4 weeks, suggesting that the measure is reliable over time and that differences caused by SCNP do not improve on their own in the absence of treatment.
The ability of the brain to integrate sensory inputs is affected in those with
recurrent neck pain.
The response time to visual and combined audiovisual inputs was slower in
the recurrent neck pain group at baseline and when measured again after 4 weeks.
Altered sensory input from the neck appears to interfere with the ability
to integrate inputs from other sensory stimuli.
Hogg-Johnson, S, van der Velde, G, Carroll, LJ et al.
The Burden and Determinants of Neck Pain in the General Population: Results of the
Bone and Joint Decade 2000–2010 Task Force on Neck Pain and Its Associated Disorders
Spine (Phila Pa 1976). 2008 (Feb 15); 33 (4 Suppl): S39–51
Lee, HJ, Nicholson, LL, and Adams, RD.
Cervical range of motion associations with subclinical neck pain.
Spine. 2004; 29: 33–40
Lee, HJ, Nicholson, LL, Adams, RD, and Bae, SS.
Proprioception and rotation range sensitization associated with subclinical neck pain.
Spine. 2005; 30: E60–E67
Lee, HY, Wang, JD, Yao, G, and Wang, SF.
Association between cervicocephalic kinesthetic sensibility and frequency of subclinical neck pain.
Man Ther. 2008; 13: 419–425
Haavik-Taylor H, Murphy B.
Cervical Spine Manipulation Alters Sensorimotor Integration:
A Somatosensory Evoked Potential Study
Clin Neurophysiol. 2007 (Feb); 118 (2): 391–402
Byl, NN and Melnick, M.
The neural consequences of repetition: clinical implications of a learning hypothesis.
J Hand Ther. 1997; 10: 160–174
Haavik-Taylor, H and Murphy, B.
Transient modulation of intracortical inhibition following spinal manipulation.
Chiropr J Aust. 2007; 37: 106
Gogia, PP and Sabbahi, MA.
Electromyographic Analysis of Neck Muscle Fatigue in Patients with Osteoarthritis of the Cervical Spine
Spine (Phila Pa 1976) 1994 (Mar 1); 19 (5): 502–506
Bränström, H, Malmgren-Olsson, E, and Barnekow-Bergkvist, M.
Balance performance in patients with whiplash associated disorders and patients with prolonged musculoskeletal disorders.
Adv Physiother. 2001; 3: 120–127
Falla, D, Bilenkij, G, and Jull, G.
Patients with chronic neck pain demonstrate altered patterns of muscle activation during performance of a functional upper limb task.
Spine. 2004; 29: 1436–1440
Paulus, I and Brumagne, S.
Altered interpretation of neck proprioceptive signals in persons with subclinical recurrent neck pain.
J Rehabil Med. 2008; 40: 426–432
Haavik, H and Murphy, B.
The Role of Spinal Manipulation in Addressing Disordered Sensorimotor Integration and
Altered Motor Control
J Electromyogr Kinesiol. 2012 (Oct); 22 (5): 768–776
Haavik-Taylor H, Murphy B.
The Effects of Spinal Manipulation on Central Integration of Dual Somatosensory Input
Observed After Motor Training: A Crossover Study
J Manipulative Physiol Ther. 2010 (May); 33 (4): 261–272
Haavik-Taylor H, Murphy B.
Altered Central Integration of Dual Somatosensory Input After Cervical Spine Manipulation
J Manipulative Physiol Ther. 2010 (Mar); 33 (3): 178–188
Ernst, MO and Banks, MS.
Humans integrate visual and haptic information in a statistically optimal fashion.
Nature. 2002; 415: 429–433
Hogendoorn, H, Kammers, MP, Carlson, TA, and Verstraten, FA.
Being in the dark about your hand: resolution of visuo-proprioceptive conflict by disowning visible limbs.
Neuropsychologia. 2009; 47: 2698–2703
van der Kooij, K, Brenner, E, van Beers, RJ, Schot, WD, and Smeets, JB.
Alignment to natural and imposed mismatches between the senses.
J Neurophysiol. 2013; 109: 1890–1899
Bresciani, J-P, Dammeier, F, and Ernst, MO.
Tri-modal integration of visual, tactile and auditory signals for the perception of sequences of events.
Brain Res Bull. 2008; 75: 753–760
Oie, KS, Kiemel, T, and Jeka, JJ.
Multisensory fusion: simultaneous re-weighting of vision and touch for the control of human posture.
Cogn Brain Res. 2002; 14: 164–176
Haavik, H and Murphy, B.
Subclinical Neck Pain and the Effects of Cervical Manipulation on Elbow Joint Position Sense
J Manipulative Physiol Ther. 2011 (Feb); 34 (2): 88–97
Laurienti, PJ, Burdette, JH, Maldjian, JA, and Wallace, MT.
Enhanced multisensory integration in older adults.
Neurobiol Aging. 2006; 27: 1155–1163
Holt, KR, Haavik, H, Lee, ACL, Murphy, B, and Elley, CR.
Effectiveness of Chiropractic Care to Improve Sensorimotor Function Associated With Falls Risk
in Older People: A Randomized Controlled Trial
J Manipulative Physiol Ther. 2016 (May); (39) 4: 267–278
Zampini, M, Brown, T, Shore, DI, Maravita, A, Roder, B, and Spence, C.
Audiotactile temporal order judgments.
Acta Psychol. 2005; 118: 277–291
Cappe, C, Morel, A, Barone, P, and Rouiller, EM.
The thalamocortical projection systems in primate: an anatomical support for multisensory and sensorimotor interplay.
Cereb Cortex. 2009; 19: 2025–2037
Meredith, MA and Stein, BE.
Visual, auditory, and somatosensory convergence on cells in superior colliculus results in multisensory integration.
J Neurophysiol. 1986; 56: 640–662
Reinoso-Suarez, F and Roda, JM.
Topographical organization of the cortical afferent connections to the cortex of the anterior ectosylvian sulcus in the cat.
Exp Brain Res. 1985; 59: 313–324
Hairston, WD, Laurienti, PJ, Mishra, G, Burdette, JH, and Wallace, MT.
Multisensory enhancement of localization under conditions of induced myopia.
Exp Brain Res. 2003; 152: 404–408
Laurienti, PJ, Kraft, RA, Maldjian, JA, Burdette, JH, and Wallace, MT.
Semantic congruence is a critical factor in multisensory behavioral performance.
Exp Brain Res. 2004; 158: 405–414
Divided attention: evidence for coactivation with redundant signals.
Cogn Psychol. 1982; 14: 247–279
Meredith, MA and Stein, BE.
Spatial factors determine the activity of multisensory neurons in cat superior colliculus.
Brain Res. 1986; 365: 350–354
Nemitz, JW, Meredith, MA, and Stein, BE.
Temporal determinants of multimodal interactions in superior colliculus cells.
Soc Neurosci Abstracts. 1984; 10: 298
The assessment and analysis of handedness: the Edinburgh inventory.
Neuropsychologia. 1971; 9: 97–113
Vonkorff, M, Ormel, J, Keefe, FJ, and Dworkin, SF.
Grading the severity of chronic pain.
Pain. 1992; 50: 133–149
Baarbé, JK, Holmes, MWR, Murphy, HE, Haavik, H, and Murphy, BA.
Influence of subclinical neck pain on the ability to perform a mental rotation task: a 4-week longitudinal study with a healthy control group comparison.
J Manipulative Physiol Ther. 2016; 39: 23–30
Juan, C, Cappe, C, Alric, B et al.
The variability of multisensory processes of natural stimuli in human and non-human primates in a detection task.
PLoS One. 2017; 12: e0172480
Haavik, H and Murphy, B.
Subclinical Neck Pain and the Effects of Cervical Manipulation on
Elbow Joint Position Sense
J Manipulative Physiol Ther. 2011 (Feb); 34 (2): 88–97
Harvie, DS, Broecker, M, Smith, RT, Meulders, A, Madden, VJ, and Moseley, GL.
Bogus visual feedback alters onset of movement-evoked pain in people with neck pain.
Psychol Sci. 2015; 26: 385–392
Return to SUBLUXATION