Research Review By Dr. Ceara Higgins©


Download MP3

Date Posted:

December 2018

Study Title:

Association of subclinical neck pain with altered multisensory integration at baseline and 4-week follow-up relative to asymptomatic controls


Farid B, Yielder P, Holmes M, Haavik H & Murphy BA

Author's Affiliations:

University of Ontario Institute of Technology, Toronto, Canada; Brock University, Ontario, Canada; New Zealand College of Chiropractic, Auckland, New Zealand.

Publication Information:

Journal of Manipulative and Physiological Therapeutics 2018; 41: 81-91.

Background Information:

Yearly, 30-50% of people experience neck pain, with many individuals experiencing what is referred to as ‘subclinical neck pain’ (SCNP), indicating lower grade dysfunction with recurrent flare-ups of pain which often go untreated (1). Several studies have shown altered proprioceptive and neuromuscular function in people with SCNP (3) which is caused by altered afferent input and plastic changes in the central nervous system (CNS) due to repetition and overuse. This type of prolonged, suboptimal input can lead to altered sensorimotor integration (SMI), altered motor output, and impairments in motor control (2). It has also been hypothesized that patients with SCNP may process visual and auditory inputs differently, leading to altered multisensory integration (4).

Generally, we utilize multiple sensory inputs in order to form accurate responses, however, when one sense is less reliable or is providing contradictory information, combining stimuli may not enhance accuracy. So, although we know that proprioception is impaired in SCNP, it is unclear how this affects the integration of other senses.

Multisensory integration is often tested using a 2-alternative, forced-choice discrimination task with semantically congruent, redundant, multisensory stimuli (5). However, when designing this type of task, it is important to consider temporal and spatial factors of the equipment, as timing and location of stimuli can reduce multisensory activity (6).

The purpose of this study was to examine differences in multisensory integration and uni/multisensory response times in patients with SCNP compared to asymptomatic individuals over a 4-week interval with no treatment given for the SCNP (essentially, they wanted to see if any impairment present was stable over the course of time). The authors hypothesized that participants with SCNP would demonstrate slower response times for both uni- and multisensory conditions due to the ongoing effects of unreliable proprioceptive feedback from the neck.

Pertinent Results:

The SCNP group showed larger variation than controls in all stimulus conditions except week 4 visual. Response times for auditory stimulus were slower in the SCNP by an average difference of 24 milliseconds (ms) at baseline and 17ms at week 4, but this difference did not reach significance. Response times for visual stimulus were slower in the SCNP group by an average difference of 48ms at baseline and 37ms at week 4 (this result did reach significance). Finally, the response times for multisensory stimulus were slower in the SCNP group by an average of 52ms at baseline and 47ms at week 4, with significance reached again.

Both groups showed faster average response times to visual stimulus than auditory stimulus. The control group showed faster response time to multisensory stimulus than to either unisensory stimuli, while the SCNP group showed fastest response times to visual stimulus, followed by multisensory stimulus, and finally, auditory stimulus. The asymptomatic group showed faster response times to all stimuli with this difference reaching statistical significance for visual and multisensory stimuli. Previous studies of individuals with SCNP have shown no differences in simple response time in these people, so these differences are not thought to be due to slower movement times in general in this population.

The authors expected to see greater multisensory gains in this study. Instead, response times for visual stimuli were almost the same as for multisensory stimuli. The smaller than expected multisensory gains may be due to a lack of complexity in the visual and auditory stimuli and the higher than expected visual response times. These higher visual response times are thought to be due to the careful attention the authors placed on technical factors in order to minimize latencies of stimulus presentations (5).

There is evidence that treatment of SCNP can improve somatosensory processing and elbow joint position sense (8) and that changing visual feedback can alter the amount of pain-free neck rotation in individuals with chronic neck pain, which may indicate an increased reliance on visual input in these patients. It is problematic if patients with neck pain are more reliant on vision, and they also have impaired visual and multisensory processing response time. Future research on the effect of spinal manipulation in patients with SCNP and the effect of treatment on response times to unisensory and multisensory stimuli could be useful.

Clinical Application & Conclusions:

Individuals with SCNP showed slower visual and multisensory response times than controls. These differences persisted over 4 weeks, suggesting that the measure is reliable over time and that the differences caused by (or at least associated with) SCNP will not improve on their own without treatment. This tells us that the ability of the brain to integrate sensory inputs is affected in individuals with recurrent neck pain. This is likely a result of altered sensory input from the neck interfering with their ability to integrate inputs from other sensory stimuli.

EDITOR’S NOTE: I can’t help but wonder what the real-world impact of these differences may be. 52 ms is certainly not a long delay – just saying! This may be a case of statistical significance not being an appropriate proxy for clinical significance? The in-depth study of the intricate relationship between neck pain (and function) and sensorimotor integration is in the early stages…but is evolving quickly! It represents a very interesting aspect the care of patients with neck pain (whether in general, from trauma or even concussion) and it will be exciting to see what emerges in this line of work in the years to come. I’ve had clinical success with manual therapy (including SMT) and rehabilitation for complex cases of chronic neck pain, headache and post-concussion syndrome patients – the patients who, in addition to their neck pain and headaches, are experiencing other symptoms like dizziness, unsteadiness, coordination issues etc. This is certainly a complex relationship and area to study, which is why I leave it to the experts!

Study Methods:

Patients were recruited from a Canadian university campus through advertisements and word of mouth. All participants were first given the Edinburgh handedness scale (7) to determine hand dominance, as the study protocols were designed for right handed individuals. The authors used this to ensure that the neck pain and control groups had similar numbers of left handed and ambidextrous participants to eliminate differences in movement or processing speed due to handedness. A screening interview was performed to remove patients based on the following exclusion criteria:
  • Recent (past 5 years) history of epilepsy
  • Recent (past 5 years) history of head injury with loss of consciousness
  • Stroke
  • Brain surgery
  • Parkinson disease
  • Attention-deficit/hyperactivity disorder
  • Psychiatric disorders other than treated depression
  • Diabetes
  • Serious vision problems (other than refractive errors or astigmatism)
  • Colour blindness
  • Uncorrected hearing problems that might affect their capacity to perform the experiment
  • History of whiplash
In order to be diagnosed with SCNP, patients had to have not sought any treatment as of baseline and were further required not to seek treatment until after the 4-week measurement session. Testing sessions were performed on days where participants had minimal to no neck pain in order to ensure that slower movements due to pain would not be a confounding factor in response time measurements. This was determined using a visual analog scale (VAS) on testing days. General neck pain severity was assessed using a chronic pain grading scale which asked patients to provide the history, frequency, duration, location, and severity of pain that day and during the previous 6 months.

This trial utilized a two-alternative forced-choice type discrimination task with multisensory redundancy, where a trial consisted of either a visual stimulus alone, an auditory stimulus alone, or a multisensory (visual and auditory) stimulus in a pseudorandomized order. Patients were asked to discriminate between the colours blue and red presented visually as a circle filled with that colour on a black background for 250 ms, audibly through a female voice saying the word blue or red for approximately 300 ms, or through simultaneous presentation of both stimulus modalities for 1 colour (never conflicting colours). All trials began with a fixation cross in the center of the screen for 1 second, followed by the stimulus, and then a response time screen for 8 seconds. Time from presentation of the stimulus to participant’s key press was recorded.

Participants were given a series of practice trials to begin, performed as many times as needed until the participant and researchers felt the participant understood the task and was able to perform it correctly. Participants responded by pressing “v” on the keyboard with their index finger when the saw/heard red, or “b” with their middle finger when they saw/heard blue, and were instructed to ignore the green stimulus. 43 trials of each colour (red or blue) were presented for each unisensory stimulus type (visual or auditory) and 63 trials of multisensory stimuli were presented for each colour. The green stimulus was presented for about 10% of the total trials and responses were not analyzed. Speaker volume was adjusted to fit participant comfort levels and ensure easy discrimination of words and patients were seated at a constant, comfortable distance from the monitor and speakers. Testing was performed at baseline and again after 4-weeks.

Outliers, defined as responses greater than 2 standard deviations from the average response time of each participant, were removed for the data pool. Although other studies have eliminated responses in less than 250 ms, the authors of this study did not as the participants continued to show high accuracy in this range.

Study Strengths / Weaknesses:

  • The researchers paid careful attention to technical factors in order to minimize latencies in stimulus presentation which may have helped avoid any negative effect of technological factors on the data.
  • The authors removed response times that fell greater than 2 standard deviations from each participant’s average response time, which may have affected the data analysis and led to more positive overall results.
  • Responses to auditory stimuli improved over time, which suggests that it may be necessary to perform multiple baseline tests to ensure that baseline performance has stabilized before performing a longitudinal study (this isn’t surprising, as participants would logically become better over time as they perform additional trials – a practice effect, so to speak).

Additional References:

  1. Lee HJ, Nicholson LL, Adams, RD. Cervical range of motion associations with subclinical neck pain. Spine 2004; 29(1): 33-40.
  2. Haavik-Taylor H, Murphy B. Cervical spine manipulation alters sensorimotor integration: a somatosensory evoked potential study. Clin Neurophysiol 2007; 118(2): 391-402.
  3. Gogia PP, Sabbahi MA. Electromyographic analysis of neck muscle fatigue in patients with osteoarthritis of the cervical spine. Spine 1994; 19(5): 502-506.
  4. Taylor HH, Murphy B. The role of spinal manipulation in addressing disordered sensorimotor integration and altered motor control. J Electromyogr Kinesiol 2012; 22(5): 768-776.
  5. Laurienti PJ, Burdette JH, Maldjian JA, et al. Enhanced multisensory integration in older adults. Neurobiol Aging 2006; 27(8): 1155-1163.
  6. Meredith MA, Stein BE. Spatial factors determine the activity of multisensory neurons in can superior colliculus. Brain Res 1986; 365(2): 350-354.
  7. Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971; 9(1): 97-113.
  8. Haavik-Taylor H, Murphy B. Subclinical neck pain and the effects of cervical manipulation on elbow joint position sense. J Manipulative Physiol Ther 2011; 34(2): 88-97.