Robin Baker - Human Navigation and Magnetoreception

As I write this Preface, it is 41 years, almost to the day(11 October 1976), since I herded eight students from the University ofManchester, UK, into the back of a small van, blindfolded them, drove them intothe Cheshire countryside, and asked them to point towards the University. Theevent was the first of what evolved to become known as the ManchesterExperiments: a series of tests designed to measure how the natural ability of humansto orientate and navigate during exploration compared with the ability of otheranimals. Over the next decade, thousands of people from all over the worldwould take part in the experiments spawned by this prototype.

Three years after this initial test the first manipulationsinvolving magnets as well as blindfolds were carried out. The results werepublished in 1980 in the journal Science[1] and in effect twohypotheses were advanced for further testing: (1) humans have some non-visualability that they can use to help solve problems of orientation and navigation;and (2) this non-visual ability is based, at least in part, onmagnetoreception. This publication triggered an invigorating air of hostilitythat at times over the next decade seemed to manifest as a series oftrans-global, trans-Pennine, and particularly trans-Atlantic animosities. Mostoften, this hostility simply led to a commendably heightened scrutiny ofexperimental protocol and statistical analysis in a zealous search for flaws(e.g.,[2]). On occasion, it also descended into personal defamation(see [3]) that did little credit to those concerned.

Each science-based criticism of the Manchester Experimentswas acknowledged and addressed either willingly accommodated or forcefully rebuffed(see Chapter 4). Strengthened by the process, the experimental evidence insupport of both of the initial hypotheses steadily mounted as the decadeprogressed. Perhaps most importantly, in 1987, I published a peer-reviewedpaper[4] in the journal Animal Behaviour that finally settledthe once-thorny problem of replication by others. The article showed that,despite the multiple and well-publicised protestations of failure, when allattempts at replications by others were treated as a whole their support forthe Manchester experiments was P < 0.001 for non-visual ability tonavigate and P < 0.005 for magnetoreception. Subsequent papers byothers, especially by Gai Murphy[5] and to a much smaller degree byMary Campion[6], served to reinforce the point still further.

As the decade entered its final quarter little of theManchester work was still being actively opposed on scientific grounds and thetime seemed right for a book to draw the whole subject together. It was also anopportunity to demonstrate how wide-ranging the studies at Manchester had beenand how much progress had been made in answering even some of the more detailedquestions. The time also seemed right, although I was hardly a neutralthird-party, to at least attempt an even-handed summary of the main points ofconflict that had arisen in the early aftermath of the 1980 Sciencearticle. The first edition of Human Navigation and Magnetoreception wasintended to do all of these things and was published nearly 30 years ago in1989.

The tone of the book was optimistic. It was also, I confess,tinged with a little pride at the progress my team of colleagues and studentshad managed to make in a relatively short time. At the books heart, of course,was the demonstration that, although weak, the ability to sense direction bymagnetoreception, then to use the information to help judge the direction anddistance of home during natural exploration, was real (Chapters 5-7). Much morethan this, though, had been discovered and all was elaborated through thebooks pages. For example, laboratory-based experiments backed up bytransequatorial displacements had shown the receptor to function more like aninclination compass than a polarity compass (Chapter 8). Also, sighted forcedexploration experiments, many through woodlands, showed that magnetoreceptionand vision normally worked together for navigation (Chapter 7). Either sense couldwork on its own, with some reduction in acuity, but without both sensesorientation and navigation were reduced to the level of guesswork (Chapter 7).Among other highlights, the location of the magnetoreceptor(s) had beennarrowed to the front half of the head, possibly in the eyes but at least inthe block of tissue including the backs of the eyes and the sphenoid/ethmoidsinus complex (Chapter 8). An unmeasured (but low) level of light may benecessary for the receptor to function (Chapter 8). Other experiments suggestedthat some element of this magnetoreceptor is set overnight and is resistant,but not impervious, to change during the day (Chapter 8). A strong magneticfield could produce such a change, the receptor then requiring hours to recover(Chapter 6). The data pointed to a sensor that consists of, or contains, apatterned array of ions, molecules and/or particles. This array is fixed by aviscous matrix (or something with the characteristics of a viscous matrix) intoa resistant pattern throughout the day, then repatterned or redeposited in asoftened matrix during sleep at night (or when damaged). At the time (late1980s) the generally favoured hypothesis was that this array consisted ofparticles of magnetite, but there were ongoing problems with this view and inSection 8.3.3 the case against magnetite was expounded and the expectationaired that this case would continue to grow.