Speed of Eye Movements (April 2020)

Introduction

A recent Jiscmail thread on the speed of eye movements in processing set a hare chasing in my brain and stimulated me to do some research.  I thought our readers might be interested in my reflections as a springboard to their own thoughts rather than as a definitive treatise on the subject. 

The thread began with a very interesting question from Dr Ruth Armstong ‘Does anyone know of any studies into the use of slow vs fast BLS or into the number used per set? 

This question sent me scuttling back to my original training manuals and to the various texts I have on the subject alongside seeking out and reading any research articles I could find to support the various views and practices expressed by my experienced colleagues in the Jiscmail trail.

 Like many of us, I was trained by Sandi Richman.  The training manual presents the theory behind eye movements or other dual attention stimulation as a means of ‘eliciting an Orientating (Investigatory) Response’.  The eye movements are seen as ‘disrupting working memory, decreasing vividness resulting in decreased emotionality and changes in somatically stored memory’.  Rapid eye movements (‘as fast as the client could comfortably tolerate the movement’) for a duration of around 20 seconds per set was advocated for reprocessing traumatic memories. 

We were also taught, however, that slow eye movements or hand taps in short sets (4 to 6 passes) were used for resource development and installation (RDI). Similarly, Shapiro’s updated book (2018) entitled EMDR therapy, principles, protocols and procedures, states that fast BLS is used for processing and slower and shorter sets of BLS for installation of positive resources including safe place. 

When I tracked down information on the use of bilateral stimulation (BLS) in RDI, I found there was controversy around this subject.  I didn’t find a lot of research information and most of it was supportive of the working memory theory of how EMDR works.  I found some that explored eye movement speed for processing aversive memories and one to support slow eye movements for resource development and installation.   

I’ll present my findings and try to highlight the ‘greyer’ areas of theory and their implications for practice. 

 Slow eye movements for resource development and installation 

A researcher who investigated the subject of BLS for RDI is Masaya Ichii.  She reported on her work at the EMDR neurobiology research symposium at the 15th EMDR Europe Association Conference, Edinburgh, Scotland in June 2014.  Thirty graduate students were used in the study.  They were asked to write down three positive memories on achievement.  These were installed under three different conditions: rapid eye movements [REM], slow eye movement [SEM] and eye fixed [EF]. Helpfully, Ichii quantified the speed of eye movement as1Hz for rapid eye movement and 0.2Hz for slow eye movements. Six sets were used each time.  She then used scales to measure general emotion (positive, negative and relaxed feeling), vividness of imagery, emotional intensity, and the internal- external attention ratio before and after the intervention. 

Her findings indicated that the most significant factor was the internal-external attention ratio.  The internally focused participants evidenced more positive changes than the externally focused participants.  This focus of attention was considerably more significant than the rate of eye movements used.  When Ichii then looked at the results for slow eye movements, she found that the internally attending group showed more vivid imagery, more positive feeling and less negative feeling than the externally attending group. She concluded that when we consider the role of bilateral stimulation, we must consider the importance of attention direction as this seems to be of greater significance than the actual speed of eye movement.  

These findings also potentially support the practice that Jiscmail cited as coming from Ad de Jong’s training.  Ad was reported as positing that using bilateral stimulation with positive memories degrades the associated positive affect.  Ad therefore advises using imaginal exposure/connection to resources without BLS.  Ichii’s findings don’t support the notion that BLS degrades positive affect. Rather, her findings report no change.  Whereas the connection to an internal focus and, therefore imaginal exposure, does have an effect for RDI.  I could then conclude that all the colleagues, including Laurel Parnell who advocate slow BLS or ‘tapping in’ when installing resources, have found ways to get their clients to have an internal focus.  In this case, the BLS would not have a negative effect on RDI even if they potentially didn’t have a positive one either.   

Another argument is that the practice of using slow BLS for RDI enculturates the client to the use of BLS while having a positive experience.  This potentially makes it much easier for the client when it comes to the faster BLS used in the reprocessing phase of the EMDR protocol.  

Faster eye movements for processing phases of EMDR protocol 

I couldn’t find much research into the speed of eye movements and the effect of speed on processing for patients.   A paper by van Veen et al, (2015) which Joanne Everson kindly drew our attention to on Jiscmail is one specific to speed and it has additional references that colleagues can look up.  Other research has highlighted the relative efficacy of different forms of BLS (Van den Hout et al, 2011).  

In line with much of the theory that underpins EMDR, van Veen and colleagues base their research on the theory of taxing working memory by holding a traumatic memory alongside performing eye movements. Since both memory retrieval and eye movements require working memory resources, there are fewer resources available for making the memory vivid and emotional during the processing and this reduces vividness and emotionality in future recall of the memory.  

The work cited in van Veen et al’s article was two studies. The first had 34 participants.  Five different eye movement speeds were compared alongside no eye movement.  The speed varied from 1.2Hz to 0.8Hz (1Hz is the same as one complete pass backwards and forwards in one second).  The participants were asked to perform a reaction time task and to rate image vividness alongside difficulty in retrieving an image.  

The second study had 72 participants who were asked to recalled three vivid aversive autobiographical memory images under each of three conditions: recall + fast eye movements; recall + slow eye movements; or recall only.  

The results

·       All eye movements resulted in slower response times compared to no eye movements indicating that performing eye movements taxes working memory 

·       Faster eye movements taxed working memory more than slower eye movements

·       Recall plus fast eye movements led to less emotional, less vivid and more difficult to retrieve images than recall plus slow eye movements or recall only. 

·       Recall plus slow eye movements fell consistently in between the effects of recall plus fast eye movements and recall only, but only differed significantly from recall plus fast eye movements.  

·       Faster eye movements produced lower image vividness and higher difficulty in retrieving the image compared to slower eye movements or no eye movements.  

The initial vividness of the memory did not effects reduction in emotionality with different eye movement speeds 

·       When the target memory is very vivid, faster eye movements resulted in greater decrease in emotionality than slower eye movements 

·       Less vivid target memories showed the same decrease in emotionality as the vivid memory with different eye movement speeds 

Discussion

The studies I found and cited here are helpful for us, as clinicians as we think about working with our patients.   In line with all our training, research confirms that loading working memory by performing a dual task does reduce emotional intensity of a target memory that is sustained over time.  Also in line with our training, faster eye movements are superior to slow eye movements for providing a greater taxation of working memory, which in turn produced memory images that were less vivid, less emotional and were more difficult to retrieve after the intervention.  The more taxing a dual-task is, the more a memory image degrades.  

Previous studies (Maxfield et al 20  ) found that the initial vividness of the image held by the patient had effected the efficacy of eye movement speed on reducing emotionality over time.   They concluded too little taxation of the dual task will leave too many resources available for vivid memory recall and its accompanying emotions and too much taxation of the dual task prevents the memory from being recalled.  They therefore suggested that if the initial image is very vivid, eye movement speed should be slower in order to keep sufficient working memory to recall the target memory.  

In the study van Veet et al hoped to replicate Maxfield et al’s findings.  They examined whether the eye movement intervention would be more effective if the load of the dual task is matched with the load in working memory. The hypothesis was that highly vivid memory images would benefit more from fast eye movements than slow eye movements and less vivid memory images would benefit more from slower eye movements than from fast eye movements. This isn’t what they found.   

They found ‘the more cognitively demanding the dual task, the more an aversive memory image can be modified, in that these images become less emotional, less vivid, and more difficult to retrieve’.  These findings are in line with working memory theory and show that no matter how vivid the initial memory, the higher speed of eye movement (1.2Hz) is the more effective in reducing emotionality and memory recall over time.  

Maybe these findings are helpful in supporting the notion that slow eye movements are useful for installing positive images and resources in that the much slower eye movements do not put a large load onto working memory.  However, I could find no research material to support this view.  If colleagues have information that might help us to understand this practice, please let us know.  

The authors also noted that their findings are not conclusive.  There is a possibility that the individual working memory capacity of a patient will have an effect on how much load is necessary or efficacious in terms of eye movement speed.  So we can look forward to continuing research and an unfolding story in the journals to come.  

References

Ichii, M. (2014)  Effect of eye movements in RDI (resource development and installation) procedure. EMDR neurobiology research symposium at the 15th EMDR Europe Association Conference, Edinburgh, Scotland in June 2014.   

Maxfield L., Melnyk W. T., Hayman G. C. (2008) A working memory explanation for the effects of eye movements in EMDR. Journal of  EMDR Pract Res  2(4):247–61. doi:10.1016/j.jbtep.2013.07.002

 Van Veen, S. C., van Schie, K., Wijngaards-de Meij, L.,  Little, M., Engelhard, I., and van den Hout, M. (2015) Speed matters: relationship between speed of eye movements and modification of aversive autobiographical memories. Frontiers in Psychiatry.  doi: 10.3389/fpsyt.2015.00045  

Van den Hout, M., Engelhard, I., Rijkeboer, M.M.,  Koekebakker, J., Hornsveld, H. Leer, A., Toffolo, M. B. J., & Akse, N. (2011) EMDR: Eye movements superior to beeps in taxing working memory and reducing vividness of recollections.  Behaviour Research and Therapy (49) pp 92-98