Blob Oculomotor Fundamentalism 

It had long been supposed that the exquisite coordination of the eyes' varied movements was all a result of brainstem calculations, and yet the requisite brainstem circuits (needed, eg, to enforce Listing's Law) could not be found; indeed, their existence was dubious because of the anatomic separation of horizontal and vertical control centers. As evidence mounted that certain basic kinematic functions were performed by peripheral, orbital mechanisms (extraocular muscle "pulleys"; see review by Miller 2007), opposition to the notion grew more strident and less coherent. The Smith-Kettlewell Eye Research Institute was at the center of this rearguard campaign.

Good Data Gone Bad 

A straightforward anatomic poster (presented at ARVO-2005) subsequently lost its primary author (who declined further association) and appeared in print as an anti-pulley polemic, funded by Smith-Kettlewell. The explicit aim of these folks was to set back oculomotor physiology three-quarters of a century:

McClung 2006 McClung JR, Allman BL, Dimitrova DM, Goldberg SJ (2006). Extraocular connective tissues: a role in human eye movements? IOVS vol 47, num 1, pgs 202-5.

Demer’s Active Pulley hypothesis supposes that orbital fibers and overlying connective tissues interdigitate. McClung et al misunderstand the proposal to be that orbital muscle fibers travel in "check ligaments" to the orbital wall, and finding no muscle in these ligaments, believe they have disproven Demer's theory.

McClung et al describe the relationship of orbital fibers and overlying connective tissue (their Figure 1D) as “demonstrating the check ligament blending into the orbital side of the muscle by investing collagen filaments around the peripheral (orbital) muscle fibers”. Demer et al (2000) describe their very similar Figure 7B as showing the “insertion of rectus orbital layer fibers on their respective pulleys”. A distinction without a difference. McClung’s figure is particularly clear in showing orbital fiber termination in connective tissue short of the sclera, near the pulleys, which is exactly what Demer et al propose. Thus, the data of McClung et al actually support Active Pulley models, yet perversely, leads them to doubt that extraocular connective tissues have any role at all in oculomotility, and to insist that extraocular mechanics is as "described in classic anatomic studies and books for over 70 years"!


I attempted to inject some sanity into the discussion with a detailed analysis of the status of the various pulley-related concepts, and of the interpretive errors of McClung et al, and others:

MillerJM-2007-Understanding-pulleys Miller JM (2007). Understanding and Misunderstanding Extraocular Muscle Pulleys. Journal of Vision, vol 7, num 11, art 10, pgs 1-15, https://journalofvision.org/7/11/10, doi:10.1167/7.11.10.

We review the stages through which pulley theory has developed, distinguishing passive, coordinated, weak differential, and strong differential pulley theories and focusing on points of controversy. There is overwhelming evidence that much of the eye’s kinematics, once thought to require brainstem coordination of EOM innervations, is determined by orbital biomechanics. The main criticisms of pulley theory only apply to the strong differential theory, abandoned in 2002. Critiques of the notion of dual EOM insertions are shown to be mistaken. The role of smooth muscle and the issue of rotational noncommutativity are clarified.

A Collection of Incomprehensible Theories

Infamous for idiosyncratic theories based on denying well-known phenomena (eg, ocular counterrolling with head tilt), Robert Jampel was invited to Smith-Kettlewell to speak against pulleys:

JampelRS-2007-Theory-of-ocular-kinematics Jampel RS (2007) A Theory of Ocular Kinematics Derived from the Commutativity of Eye Movements. Invited talk at The Smith-Kettlewell Eye Research Institute.
  • Jampel denies that eye rotations are non-commutative, apparently without knowing what the latter term means.
  • He insists that the connection between Listing's Law and the "half-angle rule" has not been demonstrated, apparently unaware that the connection is mathematical.
  • And, of course, he denies that EOM pulleys exist.

Misrepresentation, Bad Data & Conceptual Confusion 

A truly bizarre paper was then published by Arthur Jampolsky's group at Smith-Kettlewell:

LeeKM-2007-Sideslip-of-MR Lee KM, Lai AP, Brodale J, Jampolsky A (2007). Sideslip of the Medial Rectus Muscle during Vertical Eye Rotation. IOVS, vol 48, num 10 pgs 4527-4533.

The authors assert that the work of many labs over almost 20 years, supporting and refining the notion of EOM pulleys using many different approaches (see review in Miller 2007), is all placed in doubt by their irrelevant experiment, performed with flawed methodology, and brought to bear on straw-man models of their own invention. Their final conclusion – and I don't see any way to make intellectually honest sense of it – is to re-propose my original (and partly obsolete) pulley theory! Are the editors at IOVS paying attention to more than the authors' names?

Lee et al's confused analysis of flawed, irrelevant data immediately drew criticism, along with complaints about the damage such retrograde "research" does to science generally:

DemerJL-2008-Through-a-scanner-darkly Demer JL (2008). Through a Scanner Darkly.

Demer distinguishes 4 central errors in Lee & Jampolsky's work, and characterizes the pulley controversy as a paradigm shift in the Kuhnian sense, with Lee & Jampolsky as adherents to the old paradigm.
Klier-2008-Misrepresentations-and-confusions Klier EM & Angelaki DM (2008). Misrepresentations and Confusion in the Oculomotor Plant.

Klier & Angelaki express "dismay" over Lee & Jampolsky's "erroneous summary and critique of [K&A's] recent work", their "misquoting of [K&A's] papers", their "misrepresentation of [K&A's] experiments", their "mistaken claims" belied by "even a simple scan" of published work, and their "blatant misquotes and condemnations of correct experiments". Klier & Angelaki urge that recent "progress should not be undone" by reports such as Lee & Jampolsky's.
MillerJM-2008-False-differential-predictions Miller JM (2008). False Differential Predictions in Lee, Lai, Brodale & Jampolsky.

Jampolsky's group extracts two somewhat obscure results from MRI data, and asserts that the notion of EOM pulleys is thereby disproven. Actually, their results do not bear on the existence of pulleys at all, even if uncritically accepted, which because of fundamental methodological flaws they should not be. In the end, they argue that orbit-relative muscle paths are determined by "distributed resistance from the surrounding tissue," which is exactly the pulley theory of Miller (1989), which they claim to have disproved.

Lee and Jampolsky responded to these analyses with personal attacks on Demer and Miller, and insubstantial efforts to placate Angelaki's lab, never addressing the substance of any of the critiques:

LeeKM-2008-Reply-to-letters Lee KM, Jampolsky A (2008). Reply to the Letters on "Sideslip of the Medial Rectus Muscle during Vertical Eye Rotation".

Given the harsh tone of Klier & Angelaki's critique (well-justified though it was), compared to that of Demer's and mine, Lee & Jampolsky's decision to warmly thank Klier & Angelaki for their "many thoughtful comments, expressed with expected civility", while berating ours for "accusative impugning of motives ... unbefitting space in a reputable journal", was ingenuous and obviously strategic.

My letter was critical of the evidence and arguments in Lee & Jampolsky's paper, but devoid of the ad-hominem attacks they nonetheless accuse me of making. My challenge to Lee & Jampolsky to substantiate their allegations has, of course, gone unanswered. My letter spoke only to their scientific work and its lack of merit. Entry of such work into the scientific literature implies failure of the scientific review process at IOVS.

Considering the many fundamental scientific problems raised in the 3 critical letters, Lee & Jampolsky might have given some attention to the following issues:
  • Incoherent Argument. Ghasia & Angelaki (2005) and Klier, Meng & Angelaki (2006) proved that ocular plant mechanics can maintain Listing torsion across gaze shifts without brainstem involvement. Now that Lee & Jampolsky have altered their position, claiming to completely accept, rather than completely reject, the work of Angelaki's lab, they must explain how orbital tissues accomplish this feat. The explanation they offer us is that the "whole ensemble of muscles ... resists yet allows the evoked eye movement to follow the half-angle rule" as a result of the "passive mechanical reaction of other muscles that are actively maintaining the eye position". This is meaningless word salad.
  • Flawed MRI Data Collection. Lee & Jampolsky offer no defense of their unusual measurements of horizontal muscle paths in sagittal view, in which muscles were viewed against poorly contrasting globe and optic nerve as they drifted in and out of image planes with eye rotation and head movement, and and no explanation of their image segmentation methods. Instead, they simply assert that their post-processing was "more important to spatial resolution". The old GIGO adage would suggest instead that competent data collection is paramount.
  • Fabricated Models & Abandoned Theories. Lee, Lai, Brodale & Jampolsky (2007) make differential predictions based on a "pulley model" and a "non-pulley model", claiming to disprove the former and support the later. But the hard, localized pulley model they tested was a straw man of their own creation, unlike anything elsewhere proposed, and their non-pulley model had already been abandoned as implausible by DA Robinson and others a quarter of a century ago. Their letter of reply makes no attempt to defend the models upon which their conclusions rest.
  • Wrong Experiment. Their study was essentially the one I reported in 1989, with (deleterious) changes in methodology. As I explained, such a study could not then, and cannot now, in principle, distinguish plausible pulley from non-pulley models. Lee & Jampolsky offer no counterarguments.
  • Conceptual Confusion. The most bizarre aspect of Lee, et al (2007) is that they explain their own results by explicitly proposing that orbit-relative muscle paths are determined by "distributed resistance from the surrounding tissue," which is exactly the pulley theory I proposed in 1989, and which they claim to have disproved! Lee & Jampolsky's letter of reply does not explain how their research could refute the notion of pulleys in name while supporting it in substance.

Institutional Correlates

What, you may wonder, is the scientific environment like at an institute whose leadership devotes its resources to such regressive mischief? For the answer, see: Smith-Kettlewell Self-Destructs.