IASCL - Child Language Bulletin - Vol 25, No 1: June 2005
IN THIS ISSUE
IASCL HOMEPAGE CONTENTS
July 25-29, 2005
Dear Members of IASCL,
Re: Updating the governing statutes of IASCL
We have been asked to update the governing statutes of our association. We will be proposing four changes. These will be presented at the business meeting of the IASCL in Berlin, and will be voted on individually, with a two-thirds vote necessary for approval. We are sending them out now to give people time to think about them and to send comments for possible improvements prior to the meeting. [Current statutes can be accessed at http://atila-www.uia.ac.be/iascl/index.html - click on 'governing statutes'.]
All the best for now,
Mike Tomasello, President IASCL
Gina Conti-Ramsden, Vice-President IASCL
The CHILDES team at CMU has been busy over the last year, adding new facilities to the CHILDES programs and database. Here, I would like to provide a brief review of some of the most important new features.
WebData. Perhaps the most exciting new feature is the capacity to browse transcripts directly from the server while watching the related video. We have two ways of doing this, as noted at http://childes.psy.cmu.edu/data/viewer.html.
The easiest method is the WebData feature built right into CLAN. If you open up CLAN and then go to the Windows menu item up at the top of Mac screen or at the top of editor window on Windows, you can pull down to a button that says “WebData”. From here, you can choose to browse transcripts either on CHILDES or TalkBank. For this to work, you need a recent version of CLAN and QuickTime and you need to make sure that you do not have a firewall set up that is going to block streaming video (British universities often have such firewalls, for example).
Assuming you have everything right, please choose CHILDES and then double-click Biling and Yasmin and select 08.cha. You can then watch Yasmin and the Experimenter working at drawing pictures of the moon. I would say that both do an excellent job. Such materials are, of course, excellent for the study of learning sequences and conversational interaction.
Linkage to TalkBank. A second major change is CHILDES is on the organizational level. We are now thinking of CHILDES as one of several growing TalkBank databases. Others include AphasiaBank, ClassTalk, SCOTUS (50 years of oral arguments of the Supreme Court of the United States), MOVIN (CA corpora organized by Johannes Wagner), Conversation (Santa Barbara corpus), LIDES (code-switching), PhonBank (organized by Yvan Rose at Memorial University Newfoundland), SLX (Sociolinguistics), and SLA. Each of these additional components of TalkBank involves active participation of a particular research community. Of these various communities, CHILDES is the largest and most advanced. In this way, the data-sharing experiences of the child language community are serving as a guide for advances in these related fields.
Support for CA coding. It is now possible to transcribe in the CLAN editor in Conversation Analysis (CA) format. In the period between 2000 and 2003, we provided some weak support for CA transcription through a separate mode in the editor. However, we have now merged CA and CHAT formats into a single unified mode. Achieving this unification involved making some changes to both CHAT and CA. On the CHAT side, we no allow all of the basic symbols of CA such as ↑ for pitch rise or pairs of the superscript symbol ° for softer text. Altogether, we have added about 15 such CA characters to CHAT. The full list is at http://talkbank.org/ca/ca-chat-uni.html.
To capture the CA marking of aligned overlapped turns, we use the raised and lowered overlap markers, as in
*F: she says ⌈what eh⌉ .
*E: ⌊put raina⌋ .
To have these marks align properly, we needed to shift away from the variable width Arial Unicode Font that was the new standard for CHAT files to the FixedSys Excelsior 2 font that you can download from http://talkbank.org/ca/. To view overlap alignment correctly in CA files, you need to install this font. For some examples of files transcribed with these CA codes, please look at Mike Forrester’s corpus on CHILDES in the Eng-UK directory. You can access Mike’s files using WebData. You might try playing back 125.cha with Ella and her older sister Eva. For other corpora in CA notation, you can check for the CallFriend corpus in TalkBank under the Conversation directory or any of the materials in the TalkBank Classroom directory.
Sound Walker. In 2004, Leonid Spektor implemented a version of Jack DuBois’ Sound Walker utility inside CLAN. This utility allows you to repeatedly replay a segment of speech for transcription. Other methods for transcribing and linking to audio include Sonic CHAT, the video editor and the simple F5 function key method that lets you insert links with the space bar. All of these methods are described in the CLAN manual.
Progress with MOR. We have now fully tagged all of the English language corpora in CHILDES for part of speech on the %mor line. During the first pass, MOR inserts all possible tags for a word. In the second pass, POST decides which tags are most probably, given the local context. The POST disambiguation is 95% accurate. We are continuing to produce larger POST training sets to further improve this accuracy rate. However, this level of accuracy is already higher than that achieved by human annotators.
Refinement of the English MOR lexicon. During the progress of tagging the English corpora, we made major modifications to the part-of-speech assignments for English. The best way of understanding these changes is to review the files in the /lex directory of the English MOR tagger. Here you will find separate treatments for compounds with different components. For example compound adjectives formed from nouns with adjectives such as “knee+deep” are in one file, whereas compound adjectives composed of two adjectives such as “old+fashioned” are in another file. Communicators, rhymes, novel words, reduplications, and conventional onomatopoeic sequences are all separated into separate files by type. Altogether, we now have 62 English lexical files. The requirement that all files should pass through MOR meant that all forms in all of the transcripts had to map onto words in the lexicon. This led to a 3-year process of mapping and correcting forms, while still preserving original eye-dialect qualities. The biggest changes involved the differentiation of compound types. Researchers producing new English corpora will want to pay careful attention to these new requirements.
Progress with other MOR systems. We have now applied the Spanish MOR to three of the Spanish corpora and the Italian MOR to one corpus. Compounds are not a major problem in either Spanish or Italian. We also now have functional MOR grammars for Cantonese and Mandarin and the Japanese MOR grammar developed by Susanne Miyata is increasingly accurate. We have unconfirmed rumors regarding progress on expanding the French and German MOR grammars.
Tagging grammatical relations. Kenji Sagae and Alon Lavie of the Language Technologies Institute (LTI) at CMU have worked with me to build a system for automatic tagging of 27 grammatical relations in English CHILDES corpora. Kenji has also used the results of this tagging to provide automatic computation of IPSyn scores. He has shown that his results are close to those obtained by hand. The results of this work are published in a series of technical papers available from http://psyling.psy.cmu.edu/papers/ (Section 2 – CHILDES and TalkBank)
Local Servers. Some researchers would like to create and analyze data in CHILDES format without making it available outside of their work group. In some cases this is because of Human Subjects restrictions. In other cases, it is because data are not yet open for sharing. To help such groups use our tools locally, we now provide a set of instructions and utilities for setting up a local TalkBank server. These instructions are at http://talkbank.org/software/.
Phon and PhonBank. Yvan Rose and Greg Hedlund at Memorial University in St. Johns Newfoundland have been building a program designed to facilitate phonological and phonetic analysis of data transcribed in CHAT. Information on this project can be located at http://childes.psy.cmu.edu/phon/. Pending possible NIH funding for this initiative, we hope to create a new database on child language phonology called PhonBank.
XML Processing. Researchers familiar with the use of XML have begun to experiment with the newly available CHILDES and TalkBank XML data format. For example, Mark Johnson at Brown has used CHILDES XML data to teach the use of Python for corpus analysis. Similarly, Ignacio Moreno-Torres has used the XML format to set up a web utility that computes frequencies for his data on simultaneously signed and spoken language.
Collaborative Commentary. Prabhu Raghunathan is developing a program called TalkBank Viewer (TBViewer) that functions like CLAN’s WebData to view data over the web. TBViewer also includes functions that allow users to add commentary to the corpora. This facility is intended to support the construction of communities of analysis in areas such as Classroom Discourse, language socialization, formal and informal learning, tutorial dialog, CA, emergency medicine, and emergentist psycholinguistics. This facility is now in a very preliminary form and we hope to have it available for wider use by early 2006.
ELAN integration. Leonid Spektor is now working on integrating the ELAN system developed by the technical group at the MPI in Nijmegen directly into CHILDES. ELAN is an excellent system for aligning gestural annotation to corpora. ELAN currently accepts CHAT format. However, we are hoping to modify it to achieve a tighter integration. We do not yet have a target date for the completion of this ongoing work.
Database expansion. Over the last year, we have added 9 new corpora to CHILDES and 12 new corpora to TalkBank. We have announced each new corpus on info-childes. All of the corpora pass CHECK and the XML parser.
Let me encourage you to browse the database to examine some of these new facilities. I also hope that people will include discussions of these new facilities in classroom presentations of CHILDES methods.
Dear IASCL members,
By now all of you should have received your personal copy of TiLAR 3,Language Development across Childhood and Adolescence, edited by Ruth Berman. In case you have not, please let the IASCL Secretary, Steven Gillis, know about this (firstname.lastname@example.org). Any day now, TiLAR 4, Developmental Theory and Language Disorders, edited by Paul Fletcher and Jon Miller, should roll off the press at John Benjamins'. You can find the full information on this new book at http://www.benjamins.com/cgi-bin/t_bookview.cgi?bookid=TiLAR%204.
Copies of TiLAR 4 will be shipped out to you as soon as possible after that, and you should have received your copy by the time you leave to travel to Berlin. In case you have moved since the beginning of this year, please also let the IASCL Secretary, Steven Gillis, know, so he can arrange for your copy of TiLAR 4 to reach you at the correct address.
We are very grateful to all TiLAR editors and contributors for making this exciting series happen!
We look forward to seeing you in Berlin.
Annick De Houwer and Steven Gillis
Dear IASCL members,
This is my last issue as editor of the Child Language Bulletin, and I would like to take this opportunity to thank a number of people who have helped me put this publication together over the last three years.
I am very grateful to Eve Clark, Fred Genesee and Annette Karmiloff-Smith for sharing their views on the field and their personal experiences in the interviews they have very graciously granted me. Many thanks to Judith Johnston, Brian MacWhinney, Donna Thal and Beverly Wulfeck, for a touching personal portrait of Elizabeth Bates.
Special thanks to Evan Kidd, Simona Montanari and Barbora Skarabela for providing timely conference reports and pictures to go with them.
Thanks again to Brian MacWhinney for regular updates on CHILDES, to Mike Tomasello for his contributions as President of IASCL, and thanks to Annick De Houwer and Steven Gillis for keeping us posted on the progress of the TiLAR series.
Finally, we all have an ongoing debt of gratitude towards Joris Gillis and Steven Gillis, without whom the Bulletin would not reach you electronically.
See you all in Berlin!
Child Language Bulletin Editor
Professor Annette Karmiloff Smith is Head of the Neurocognitive Development Unit at the Institute of Child Health in London where she runs a research team looking into infant and child development. She is a leader in the field of typical and atypical infant and child development, in particular with respect to language acquisition, face processing, drawing, writing, and problem solving.
In addition to innumerable publications in leading scientific journals, Prof. Karmiloff-Smith has recently co-authored two books on language and cognitive development aimed at the general public with her daughter Kyra: Everthing your baby would ask if only he or she could speak, and Pathways to Language: From foetus to adolescent.
Ludovica Serratrice: Professor Karmiloff-Smith, you started your career at the International Centre for Genetic Epistemology in Geneva with Jean Piaget. How do you remember that period of your life, and how did it shape your research interests?
Annette Karmiloff-Smith: Geneva was a period of intense stimulation and intense frustration! Before I started psychology, I was a simultaneous interpreter working for various United Nations organisations in Geneva and was bored with merely repeating other people’s thoughts. I was reading about child psychiatry and in the bookshop often saw books about child psychology with Piaget’s photo on the cover. One day he came into the bookshop, so I followed him to the university and audited a lecture. I was bowled over by his approach and signed up for psychology that Autumn. After my licence (roughly an MSc in UK) and two years working in the Palestinian refugee camps in Beirut, I came back to Geneva University to do my doctorate and to work at Piaget’s Centre. Piaget was both inspirational and obstinate! The inspiration came from his view of human ontogeny as a vital way to understand adult knowledge and his breadth of vision of the child as a whole. The frustration came from Geneva being a “school of thought”, with Piaget being called “Patron” (boss!) and a lack of real debate. I was considered a heretic. But, although I criticised Piagetian theory in quite a lot of my writings, I realise in looking back just how much his epistemology continues to influence my thinking today. Unlike others, I have never thought of myself as, say, a psycholinguist because Piaget made me always see the child as a whole, so I have been just as interested in number, drawing, writing, reading, face processing, child physics, theory of mind, and other domains as I am in language.
LS: After a period at the Max Planck Institute for Psycholinguistics in Nijmegen, and subsequently at the MRC Cognitive Development Unit in London, during which you continued to work on typical language acquisition, you started to study predominantly atypical language and cognitive development. What prompted this shift towards developmental disorders?
AKS: The Cognitive Development Unit (CDU) was an exciting place of debate, not only about psychology, but about the philosophical issues of nativism, empiricism and constructivism. Some of the CDU scientists were pretty extreme nativists; others were constructivists. Many were working on atypical development, particularly dyslexia and autism. Those of a nativist persuasion often called on examples of so-called dissociations in atypical development to make their case. In such an atmosphere, I obviously started to read a lot about atypical development, and it was a visit to Ursula Bellugi’s lab in San Diego that finally got me involved in my first small research project on atypicality. And, Williams syndrome seemed at first blush to present a clear-cut case of independently “spared” and “impaired” modules so, with all the excitement about Fodor’s “Modularity of Mind” at the CDU, this seemed a perfect testing ground. But I soon found that the claims about spared/intact modules didn’t hold, once one probed the cognitive processes underlying the successful behaviour.
LS: One of the main themes of your ongoing research programme is the argument against the inadequacy of the static model of the adult neuropsychology literature for the study of developmental disorders. Why is it so crucial to adopt a dynamic perspective when it comes to development?
AKS: First and foremost, one has to remember that neuropsychology patients who suffer sudden trauma in adulthood (accident, stroke etc.) had developed normally until their trauma. The specialisations in their brain had already been consolidated, and thus it is possible if the brain insult is very focal, domain-specific impairments may ensue. But children with early focal brain damage or those with genetic disorders originating during embryogenesis are a different kettle of fish. Their brains are in the process of developing and we know from work by Neville, Rakic and others, that early on the different regions of the brain are highly interconnected. It is with development that they become increasingly specialised. Moreover, in young infants, a tiny impairment can have cascading effects on development, affecting some areas – because of the neurocomputational constraints on those areas - more than others. This means that brains can result in seemingly domain-specific impairments and preservations, whereas they didn’t start that way. So, the endpoint of development is not always informative about the start state of development, and yet nativist claims are about the start state in infancy, but on adult neuropsychology patients or older children and adults with genetic disorders. So my argument is that nativist arguments cannot rely on mere inferences from the end state in middle childhood or adulthood. We have to examine low-level processes in the start state, not emergent cognitive-level modules in the end state.
LS: Current advances in the sequencing of the human genome have opened up exciting new lines of enquiry in the relationship between genes and human cognition. You have, however, always urged great caution against a simplistic cause-effect relationship between genes and complex cognitive systems. Could you elaborate on your rationale for doing so?
AKS: Genes that are expressed in the brain don’t code for human cognition! Their protein products are involved in processes like developmental timing, neuronal migration, neuronal type/size/density/orientation, myelination, lamination, ratio of grey matter to white matter, firing thresholds, neurotransmitter differences, dendritic growth, synaptic regulation and so forth. Most genes whose expression we know enough about are expressed widely across the brain, not targeted to a single brain area like parietal cortex, for instance. And genes expressed in the brain are usually also expressed in various other parts of the body. There is no simple one-to-one mapping between genes and cognitive outcomes. Many-to-many mappings seem to be the rule. Even if only intended as a shorthand, expressions like a “gene (or specific set of genes) for language” or “gene (or specific set of genes) for number” are dangerously misleading. Even in syndromes with a single gene mutation, like FragileX syndrome, the gene is expressed widely and is involved in something as general as synaptic regulation. What is clear is that developmental timing plays a crucial role in emergent outcomes. To reiterate, there is no simple one-to-one mapping between genes and cognitive outcomes, so the story is going to be exceedingly complex, far more complex than the usual glib references to genes found increasingly in the psychological literature.
LS: The role of environmental factors is sometimes downplayed when it comes to the genotype-phenotype relationship. What are the different levels on which the environment plays a role in this complex relationship?
AKS: Well, it depends what one means by “environment”. First, there is first the genetic environment itself: genes interact with one another, the role of some being to switch on or switch off other genes at different points in the developmental process. Second, there’s the external environment, both social and physical, but it has to be recalled that the interaction between gene expression and environment is two-way; environmental influences on genetic expression then influence subsequent gene expression. I think for psychologists, the question is not whether both genes and environment play a role. Everyone agrees that they must. The question many raise is whether one plays a more predominant role than the other. For Nativists, the genetic side of the equation is predominant; for Empiricists, it’s the environmental side. But for Neuroconstructivists the very question is misfounded. Development is dynamic and one cannot apportion a certain percentage to genes or to environment. Behavioural genetics does, but for the variance between individuals, not for the actual trait itself. So when they claim that, say, 50% of X is due to genes, it doesn’t mean that 50% of the trait is due to genes, but 50% of the differences between individuals for that trait is due to their genetic inheritance. That’s very different from the aims of molecular genetics. Don’t forget that even monozygotic twins raised together differ in their outcomes. So, the interaction between genes and environment is much more complex than we imagine, and we are still far from an understanding of precisely how individual genes are expressed and interact with other genes and the environment. And we are still very far from a precise account of what a child’s environment is over developmental time.
LS: In the attempt to model the effects of gene deletion/over-expression in humans, scientists typically refer to animal models. Mice are commonly used in these experiments as they share 99% of the human genome, breed very rapidly, and display a high level of conservatism of many developmentally important genes. What are the advantages and the limitations of this type of research?
AKS: Animal models are often a very good first approximation, but we need to be extremely cautious in how we generalize from animal to human. For instance, there was an excellent mouse model of Williams syndrome (WS) by Meng and collaborators. The mouse carries on chromosome 5G the same genes as those deleted in WS on one copy of chromosome 7 and in the same sequence, albeit reversed. The researchers were interested in testing the hypothesis that one gene, Limkinase1 (LIMK1), was causal in the WS spatial deficits, as had been claimed by Frangiskakis and colleagues working on WS and partial deletion patients. They made a mouse knockout of LIMK1 and found that the mice displayed spatial deficits in the Morris Maze. Many took this as proof that LIMK1 was causal in the WS spatial deficits. But, with colleagues in Manchester, we identified several patients with partial deletions in the WS critical region including LIMK1 but who had no spatial impairments. So, although mouse models are excellent, they do in my view have several limitations. First, we cannot take for granted that the same gene has identical functions or identical expression timing across species; this has to be demonstrated. Second, there is a problem addressing spatial impairments, because the deletion of many different genes, not just LIMK1, causes spatial impairment. This is probably because multiple genes affect the mouse’s intelligence generally and mouse intelligence is mainly measured through spatial deficits shown by the Morris Maze. Third, WS involves the deletion of some 25 contiguous genes, not a single gene deletion like the mouse model, yet we need to know how LIMK1 interacts with other genes, say at the telomeric end of the deletion. Fourth, LIMK1 deletion affects more than spatial deficits in the mouse, and we need to compare these other deficits with WS before drawing conclusions. Fifth, what we measure in the mouse is very different from what we measure in the human: spatial navigation in the mouse, compared to table-top cognition (e.g., Block Design) in the human where the representation of one’s body in space is not important. In my lab and in recent ongoing collaborations with other groups, we have been examining spatial navigation in WS and partial deletion patients to compare more directly with the mouse model. Still we’re left with the problem of the mouse behavioural repertoire being very different to that of the human, Yet, even when animal models are much closer to the human such as the chimp, they are far from straight forward. For example, after making a chimpanzee model of Down syndrome (trisomy of chromosome 22 in the chimp), The International Chimpanzee Chromosome 22 Consortium concluded that “..the biological consequences due to the genetic differences are much more complicated than previously speculated.’ And this was by an international group of specialists as recently as 2004! I am not arguing that animal models aren’t useful. They are. I am simply cautioning against rapid generalizations to the human case.
LS: Computational models have also been used alongside animal models to understand the correlation between a gene, or set of genes, and cognitive behaviour. What have we learnt from connectionist networks, and how much further can they take us?
AKS: I believe computational models to be very useful as tools
to provide in principle demonstrations. For example, we’ve shown that lesioning
a network in its infant state versus in its adult state at the end of learning
gives rise to very different results. If noise is added to the system, the
infant network is much more impaired, probably because it requires clear inputs
to form and consolidate clear representations. Hence, the importance of motherese
to language-learning children. The adult network has already built such clear
representations, so noise affects it far less. By contrast, if a connection
is severed, then the infant network finds an alternative trajectory for learning,
whereas the adult network is seriously impaired because its trajectories have
already been fixed as a result of learning. In other work, we’ve shown that
domain specificity can emerge from domain generality as a result of processing
different inputs. Some think of connectionist models as theories. I see them
more as tools, which,, because of the need to be precise about everything,
highlight the huge conceptual leaps we risk making with discursive verbal
theories. But I find all theories more convincing when they also generate
novel predictions that can then be tested empirically.
LS: In addition to your scholarly work you have also published a number of books aimed at the general public. How do you see the role of the scientist outside the confines of the academic world?
For the most part, academic salaries come from public taxes. So, if only for that reason but for many others too, we owe it to the public to make our science intelligible to them. And the public is interested in science. I've written a couple of books, one with my younger daughter, for parents and childminders in which the approach is quite lighthearted but the science is very serious. The challenge is to be simple without being simplistic. I've also been a consultant to industry on design of toys, books, and videos, always with the aspiration that the scientific knowledge I have of infant and child development will be incorporated into new products. Recently, I have been involved in developing and bringing to life Pampers World of Babies which has been exhibited in Switzerland, Sweden, Greece and the UK. This is not strictly an exhibition, but a very interactive experience, which aims to bring to parents and other adults the experience of the world from a child's mind.
For example, adults have to learn to sit again on very uneven wobbly stools where it's difficult to find your centre of gravity, to walk along a waterbed gripping onto furniture, to walk in the park holding onto huge (father's) hands on a fast treadmill, to write with a huge pencil whose nib is fixed to a moving coil, and so forth. Everything in the experience is huge, simulating objects in the world of a tiny child. And Procter and Gamble have funded a research proposal I put to them to form a Pampers European Research Consortium (PERC) together with CNRS Paris and Max Planck Munich, to study longitudinally in infancy the relationship between mother-infant interaction and domain-specific and domain-general developmental changes in various cognitive domains like speech processing, face processing, attention, and causality - nothing to do with nappies! Yes, I've been a consultant for a nappy company (blush, blush!) but on projects like brochure writing, website article writing, the above experience, all offered free of charge to parents and that had nothing to do with nappies, but part of the company's mission to bring knowledge of child development to their customers. My next target is teenage mothers, since Britain has the highest level in the whole of Europe. So, yes, I think it is crucial to translate our scientific knowledge into forms that are useable by the general public to enhance their understanding of children's minds. I think I am a better albeit exhausted grandmother (I've 7 tiny grandchildren!!) since I've been thinking about these issues in practical terms.
LS: Professor Karmiloff-Smith, thank you very much for your time.
French narrative corpus
A new corpus of French picture sequence descriptions from Monique Vion and Annie Colas of the Université de Provence. The subjects were 7, 9, and 11 year-olds and the experimental design specifically compared logically ordered vs. arbitrary picture sequences.
Bodor, P. (2005).
On Emotions: A Developmental Social Constructionist Account.
L' Harmattan Publishers, Hungary.
The Seventh Annual International Conference of the Japanese Society
for Language Sciences
Sophia University (Yotsuya campus), Tokyo, Japan
Workshop on Computational Modeling of Lexical Acquisition - The Split
University of Split, Croatia
Psychocomputational Models of Human Language Acquisition
University of Michigan Ann Arbor
The state of the art in speech error research
LSA Institute in Cambridge, Mass.
The 30th Annual Boston University Conference on Language Development
The IASCL is a worldwide organisation, which means that is aims to serve child language researchers in all countries of the world. Child language research is important everywhere, both from a theoretical perspective (cf. for instance the significance of cross-linguistic evidence) and from a more applied point of view (cf. for instance the need for good description to allow for the assessment of language learning problems). Unfortunately financial considerations are often a hindrance to the development of scientific disciplines in countries with severe economic problems. The IASCL has always been supportive of would-be IASCL members working in such countries by waiving membership fees for them.
IASCL funds are limited, though. In the past, donations from regular IASCL members have been very helpful in supporting colleagues from economically disadvantaged countries. In order to continue offering that support, your donations are very welcome indeed. Each donation, whatever the amount, will be acknowledged by a receipt signed by the IASCL Treasurer (useful perhaps for tax purposes). You may send donations in either pounds sterling or American dollars.
(1) Cheques in pounds sterling payable to IASCL can be sent to:
Dr Anna Theakston
University of Manchester
Department of Psychology
Manchester M13 9PL
Cash payments in pounds sterling can also be made by prior arrangement with Dr Theakston (email@example.com) at the above address.
(2) For American dollar amounts, please send your donations to:
Prof. Judith Becker Bryant
IASCL Assistant Treasurer
Department of Psychology, PCD 4118G
University of South Florida
Tampa, FL 33620-7200
The IASCL as a whole will be sure to benefit from the more diversified nature of its membership as a result of your donations. Many thanks in advance!
Anna Theakston, IASCL Treasurer