Obituary for Dr. Marian Annett, 1931–2018

Marian Annett in the robes of D.Sc. of the University of London

Marian Annett (née Drabble) was born in 1931 in the small Derbyshire mining village of Whitwell. Her father, Harold, who was an army runner in the First World War, and had received the Military Medal, worked on his father’s farm, while Marian’s mother, Joyce, worked in the village shop.

Marian decided early to study psychology at university, and a family story says that she may have been influenced in her choice of subject by Alfred Hitchcock’s 1945 film Spellbound, in which the beautiful but emotionally cold psychiatrist played by Ingrid Bergman attempts to unravel the story of an amnesic patient masquerading as a doctor. However, psychology could only be studied at university if science had been studied at school, and Marian only managed that by persistence, persuasion and by changing schools to Shirebrook Grammar School for Girls. While she was there, Marian developed an abiding interest in music and playing the piano, and in 1947, aged 16, she became an Associate member of the London College of Music.

In 1949 Marian entered Bedford College, a women’s college at the University of London, graduating in 1952 with a first-class honours degree in psychology. After graduating, Marian studied clinical psychology at the eminent Maudsley Hospital in south London, being awarded a Diploma in Abnormal Psychology in 1953, which she often said stood her in good stead in testing children.

When applying for a post at the Burden Neurological Institute in Bristol in 1954, Marian met John Annett, who was also applying for a position. Fortunately, two jobs were available, both applicants were appointed, and John subsequently became Marian’s husband. They wed in 1955 and were married for 63 years.

In 1956 Marian moved to Oxford as a clinical psychologist at the Warneford Hospital where she remained until 1960. After their time at Oxford, Marian and John spent a year at Port Washington, Long Island, USA before returning to the UK in 1961 to work in Sheffield, moving to Aberdeen in Scotland in 1963 and later (1965) to Hull. While at Hull (where she held an honorary research position from 1965 to 1972) and as an external candidate of the University of London, Marian submitted in 1965 her PhD thesis entitled, “Some aspects of the growth of some common concepts in children and adults”. The thesis was firmly in the Piagetian tradition and included data from Marian’s first published paper (Annett, 1959).

Marian subsequently had an honorary research fellowship at the Open University (1973–1974), and then, beginning in 1975, worked for several years in Coventry at Lanchester Polytechnic (now the University of Coventry).

In 1987 Marian moved to the University of Leicester where she was appointed Lecturer, being promoted in 1993 to Reader in Psychology. In 1992 she was awarded the degree of D.Sc. of the University of London. At this time John Annett was Professor of Psychology, eventually becoming Emeritus Professor, at Warwick University, and he and Marian lived on the edge of the Warwick campus. Marian worked at Leicester until her formal retirement in 1994, then becoming Honorary University Fellow/Reader Emerita.

To mark the occasion of Marian’s retirement, Alan Beaton organized a symposium at the British Psychological Society’s 1995 meeting held at Warwick University, with contributions from Marian herself, Tim Crow, Dorothy Bishop and Chris McManus. After the symposium, Marian and John invited the symposiasts to their house for tea and cakes, and a pleasant couple of hours were spent chatting informally about laterality and other matters. For Alan, this was only one of many such occasions over a number of successive years. While working at Open University summer schools based at the Warwick campus, he would visit Marian and John who invariably had tea, sandwiches and Marian's latest data, and whatever paper she was working on, laid out for discussion when he arrived. After the academic talk had ended, the conversation would turn to the latest news of their children in whom Marian and John showed an obvious but modest pride.

After retiring, Marian continued to write for many years, her last paper (Annett, 2011) being published in 2011 (see list of publications). During retirement, Marian and John travelled widely, usually on cruises, and maintained their interest in music and the theatre. Marian continued to play the piano and was an avid reader. Eventually, her eyesight and her general health declined but she continued to take great delight in her family, especially her grandchildren. She died on 27th April, 2018 after a period of time in a care home and hospital. She is survived by her husband, Professor John Annett, her children Professor James Annett and Dr. Lucy Annett, her grandchildren, and her sister, Dr. Mary Bonsall.

Marian’s life-long interest in laterality began with a book chapter in 1961, arising from her clinical position in Oxford. Marian and her colleagues suggested that mixed-handed children with unilateral epileptic foci tended to have verbal functions subserved by the cerebral hemisphere in which the focus was located, in order that spatial functions could be carried out by the intact hemisphere. The relationship between hand use and cerebral lateralization, that earlier had been studied by Marc Dax, Paul Broca, John Hughlings Jackson, Pierre Marie and Oliver Zangwill, would occupy Marian’s research interests almost exclusively for the rest of her academic life.

“There are many puzzles about left handedness” were the opening words of a key paper that Marian published in Nature, entitled “A model of the inheritance of handedness and cerebral dominance” (Annett, 1964). Marian pointed to the lower consistency of handedness in left-handers than in right-handers, a higher likelihood of recovery from aphasia, and greater prevalence in some pathological groups. Although five decades later some may take issue with those conclusions, they were undoubtedly important issues to be explored and were mostly little researched at the time.

As the title of the Nature paper suggested, genetics was hypothesized to underpin variation in lateralization. A model was proposed with two alleles, D “dominant” and R “recessive”, in proportions 0.8 : 0.2. DD individuals were right-handed (with language mostly in the left hemisphere), RR individuals were left-handed (with language mostly in the right hemisphere), and DR heterozygotes showed partial penetrance with about 25% being left-handed and with language in either hemisphere. As Marian later wrote, this model is a “variation on the classic gene pair theme” (Annett, 1985, p. 53). The model predicted that the children of two left-handed parents should be left-handed. Collecting large numbers of offspring of two left-handed parents is not easy, and Marian’s important 1974 paper, “Handedness in the children of two left-handed parents”, showed clearly (Annett, 1974) that handedness does not “breed true”, fewer than fifty per cent of the children of two left-handed parents being left-handed, which was a key empirical finding. That may have disproved the 1964 genetic model but it opened the way to the right shift (RS) theory.

Marian had always been struck, as noted in the Nature paper, that many people are not consistently right- or left-handed, using their right hand for some actions and their left for others. Failure to separate such “mixed-handers” from consistent right- or left-handers was, Marian believed, one reason for disagreement in the literature on how handedness related to hemispheric language dominance. This issue determined the course of her career. Measuring handedness became a central feature of Marian’s work, both in terms of preference and in terms of performance. Her paper published in 1967 (Annett, 1967) used questionnaire items gleaned from earlier questionnaires by Humphrey, and Crovitz and Zener, resulting in the Annett Hand Preference Questionnaire (AHPQ), used in all of Marian’s later research. Using association analysis, “to describe the main types of hand preference so that they could be defined on empirical rather than a priori grounds (p.303)”, Marian subsequently described the great variety of combinations of hand use (Annett, 1970), with 24 different patterns from consistently right-handed to consistently left-handed with no obvious discontinuities between adjacent categories of handedness. This was the origin of her view that the distribution of hand preference is essentially continuous and approximately J-shaped (Annett, 1970).

Unlike other questionnaires, the AHPQ is not scored by summing across items but is used to identify a number of categories, or hand preference classes, with participants originally allocated to one of eight classes, later reduced to seven. However, few other researchers have adopted this approach.

Marian’s 1970 paper presented the peg-moving task devised in collaboration with her husband John, whose expertise was in motor skills. The Annett pegboard, still used extensively in handedness research, showed mean differences in skill between the hands for participants in the various hand preference classes. The pegboard data showed a normal distribution of between-hand differences and reconciled the J-shaped distribution of hand preference with the normal distribution of between-hands skill differences. From then on the normal distribution of skill asymmetries provided the basis for Marian’s theorizing.

The essentials of her RS theory were set out in 1972 in a paper (Annett, 1972) that at the time received little attention. An updated version of the theory was later published privately (Annett, 1978). A key insight of Marian’s was that chance played an important role in determining laterality, some factor coding for presence versus absence of asymmetry. The essentials of the right shift (RS) theory as set out in these two papers are as follows.

• Genes do not determine handedness but instead determine language lateralization. Some individuals have a “right shift” (RS+) and others do not (RS−).¹ RS+ individuals all show left-hemisphere language dominance, whereas language dominance in RS− individuals is a matter of chance, 50% right and 50% left. The alleles determining RS+ and RS− are rs+ and rs−, with rs+ dominant; rs++ and rs+− genotypes are RS+ and the rs−− genotype is RS−. In a large series, 9.27% of individuals with dysphasia after unilateral lesions had right-sided lesions and must represent half of the RS− individuals and hence half of the rs−− individuals. On that basis, the rs− allele has a frequency of √(2 × 0.093) = 0.4306, rs+ has a frequency of 0.5694, and hence the three genotypes rs++, rs+− and rs −− have frequencies of .3294, .4904 and .1854 respectively (Annett, 1978).

• Handedness depends on the extent of the right shift, RS‐ individuals having a normal distribution of between-hands skill differences centred at zero (and a standard deviation of one), whereas RS+ individuals have a normal distribution with a standard deviation of one but with a mean at some positive value, estimated at 1.937 standard deviations to the right. These distributions are for asymmetry of skill. Whether or not a person is right- or left-handed for preference depends on a threshold or cut-off dividing the overall distribution which may be affected by social or other pressures. Those to the right of the threshold are right-handed and those to the left are left-handed. (In a later development of the theory Marian suggested that the rs+ gene operates in a dominant-recessive mode in the determination of cerebral asymmetry for language, but in additive fashion in regard to handedness.)

• The 1978 paper also considered handedness in twins, and for that group, a lesser right shift has to be invoked in order to explain the data. The paper also speculates that there may be a sex difference in the extent of right shifts.

• The paper finishes by considering what may be maintaining the genes for handedness, and it is pointed out that the frequency of the rs+‐ heterozygote is about 49%, and therefore maximal, “suggesting that the gene is most advantageous in single dose” (p. 18).

Implicit in the RS theory (but see below) is the suggestion that the right shift is only present in humans, and that non-human animals only show the RS‐ phenotype. Handedness in apes and primates had then been little studied, and Marian and John spent long hours at Twycross Zoo in Leicestershire observing the hand used by 31 gorillas when spontaneously reaching for food. Preferences varied from strong left to strong right, “most animals showing intermediate levels of preference (p. 274)”, with individual animals being consistent in their preference (Annett & Annett, 1991). Apparently, Marian spent so long observing them that the gorillas recognized her when she visited the zoo.

A major summary of Marian’s right shift theory appeared in her book Left, Right, Hand and Brain: The Right Shift Theory (Annett, 1985), which reported further data and modelling, particularly developing the idea of a balanced polymorphism plus heterozygote advantage (BP + HA). Across the population as a whole, therefore, the rs+ gene might entail costs as well as benefits.

Marian had long been interested in dyslexia, and her review of 25 studies, showing a clear excess of non-right handedness in people with dyslexia, is the first meta-analysis of the relationship (Eglinton & Annett, 1994). A later development of the RS theory suggested that advantages of the rs++ genotype in relation to speech and language functions in the left hemisphere arise at the cost of slight impairment to right hemisphere functions (Annett, Eglinton & Smythe, 1996), with perhaps there being two kinds of developmental dyslexia associated with different handedness patterns (Annett, 2011).

The RS theory was developed and extended further in two later works. The major theoretical publication during this period was a paper in Cahiers de Psychologie Cognitive/Current Psychology of Cognition (CPC), presented as a target article with twenty commentaries and Marian’s responses (Annett, 1995a, 1995b). Together the article and commentaries provide a good standpoint from which to view perceptions of the right shift theory, twenty years after it had been enunciated.

Marian’s final major publication was the book Handedness and Brain Asymmetry: The Right Shift Theory, published in 2002. In part a revision of Left, Right, Hand and Brain, the new book was also a progress report on Marian’s more recent research, particularly, “A search for evidence relevant to a BP + HA for the RS+ gene [which] has been the main focus of my research for over 20 years” (Annett, 2002, p. 186). The research endeavour ranged across a broad spectrum of topics: motor function, reading and dyslexia, mathematical reasoning, spatial ability, vocabulary development, language learning, sport, and handedness in veterinary surgeons.

Marian had been interested in the relationship between handedness and schizophrenia since at least 1985, when she mentions it in Left, Right, Hand and Brain. In 1997 she presented a more extended theory, summarized in the paper’s title, “Schizophrenia and autism considered as the products of an agnosic right shift gene” (Annett, 1997). Once again the target article was presented with commentaries and a response. Tim Crow, who had long argued that schizophrenia resulted from a disorder of mechanisms leading to cerebral asymmetry, discussed Marian’s 2002 book in a review with the striking title, “What Marian Annett can teach Noam Chomsky and could have taught Stephen Jay Gould if he’d had time to listen”, concluding that “the gene predisposing to psychosis was the gene for cerebral asymmetry, i.e., Annett’s right-shift factor” (Crow, 2004, p. 124). Marian did however disagree (see Annett, 2000) with Crow’s theory that cerebral asymmetry was “the speciation event” distinguishing Homo sapiens from previous species since, according to the RS theory, a high proportion of individuals do not carry the right shift gene, being of the rs−− genotype. Furthermore, she applied her theory to the analysis of handedness in chimpanzees, arguing that they show a small shift to the right, but of much smaller magnitude than that for humans (Annett, 2006). If so, it raises the possibility that the right shift factor was present in the common ancestor of humans and chimpanzees more than six million years ago.

Although Marian sought advice from experts in other fields whenever she felt the need, almost all of her many research papers were published under her sole name. Apart from joint papers with her husband John, and those co-authored with her research assistants, her scholarly output was achieved without the international collaborative effort which today would be typical of a world-class researcher at the forefront of her field. However, her ideas struck a chord with a North American neurologist with an interest in cerebral dominance, Dr. Michael Alexander. They jointly published in 1996 a pair of linked papers testing predictions from the RS theory as to the relative proportions of atypical hemispheric speech representation expected in right and left-handers for different criteria of sinistrality (Alexander & Annett, 1996; Annett & Alexander, 1996).

Quiet, indeed reserved, Marian was not one of those academics who trumpet their own achievements at every opportunity while denigrating those of others. In defending the RS theory, she let reasoned argument and her elegant writing speak for itself rather than engage in the kind of blatant self-promotion sometimes seen at academic meetings. Nor was she an unsung hero. A prize possession on her desk was a photograph of the left-handed tennis player, Jimmy Connors, signed by him to “Dr. Marian” after hearing a radio talk in which he was used as a counter-example to the myth that all left-handers are clumsy.

How posterity will regard the RS theory is unclear, but it undoubtedly was influential in stimulating a host of important empirical and theoretical questions about familial patterns of handedness and language dominance, questions which still need exploring as much today as they did when Marian published her 1964 paper in Nature. That is a legacy of which any scientist can be proud.

Acknowledgements

We are grateful to Marian’s family for their help in writing this obituary, and in particular to Professor James Annett for allowing us to use details from the eulogy he wrote for Marian’s funeral. We are also grateful to Dr. Tabea Cornel for useful discussion and assistance.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Marian’s writings do not always distinguish between right shift as a phenotype and right shift as a gene. Here we use RS− and RS+ to indicate the phenotype, and rs− and rs+ to indicate alleles, with genotypes rs−−, rs− + and rs ++.