[numcog] Re: Greetings all!
- From: Edward Hubbard <edhubbard@xxxxxxxxxxxx>
- To: numcog@xxxxxxxxxxxxx
- Date: Sun, 20 Apr 2003 15:44:23 -0700
Hi All,
To introduce myself, I am a fourth year PhD student at the University of
California, San Diego. My work is perhaps unusual for list in that (to
date) I have not been exploring numerical cognition directly. Rather, I
study a group of people who experience synesthesia, an unusual mingling of
the senses. For example, some people will report that they see colors when
viewing letters and numbers, while others will report colors, and spatial
forms when thinking of the months of the year, the days of the week or
sequences of numbers. Others report seeing colors for music, or words
(auditorily presented). Even rarer forms involve people feeling tactile
shapes when tasting something or when listening to music, or even tasting
food from hearing words. My specific research has focused on a group of
synesthetes that report seeing colors when looking at letters and/or numbers.
There are four defining characteristics of synesthesia: 1) it is automatic
2) it is involuntary, 3) it is consistent (i.e., reproducible) and 4) it is
systematic. These four characteristics can be seen in all synesthetes,
while other characteristics (such as having a spatial extent, or an
emotional connotation) are seen in only some synesthetes.
Research back as far as 1880 has demonstrated that synesthesia runs in
families, and is more common in women than it is in men, suggesting that it
might be a genetic condition, with an x-linked dominant mode of inheritance
(although the loci studies have not yet been carried out). More recent
research has demonstrated that, although each synesthete reports
experiencing idiosyncratic colors, synesthetes are quite consistent over
test-retest intervals of more than one year. Additionally, as you might
expect given that synesthesia is automatic and involuntary, synesthetes
show Stroop-like interference if they are presented with "miscolored"
numbers (for example, if a 2, experienced as green is presented in red ink).
My early research was based on the attempt to prove that reports of
synesthesia were not only veridical reports, but were reports of genuine
sensory states, and to show that the synesthetically experienced colors
were able to influence performance on psychophyscial tasks. For example,
we presented synesthetes and non-synesthetes with displays composed of
numbers (e.g., a field of 2s) and embedded a figure composed of (e.g.) 5s
(either a square, rectangle, diamond, or triangle) in the display. For
non-synesthetes, this task is extremely difficult, as the 5s are mirror
images of the 2s. On the other hand, for a synesthete who sees (say) 5s as
red and 2s green, this is seen as a red triangle against a green
background, and, accordingly, pops-out.
In a second experiment, we took advantage of a phenomenon known as
crowding. If a single stimulus is presented in the periphery, it is easy
to identify. However, if presented in the presence of other, flanking
items, the target stimulus becomes much more difficult to
identify. Previous research had shown that the magnitude of the effect is
reduced when the target and flankers are presented in different colors. We
reasoned, that, if synesthetic colors were truly sensory, they should, like
real colors, reduce the magnitude of the crowding effect. We found that,
indeed they do. However, the most interesting thing about this was the
reports of our synesthetes when we asked them afterwards about the
experiment. They reported, "Although I couldn't see the number, I saw red,
so I know it must have been a five" or (for a synesthete who sees colors
for letters) "I couldn't see the letter, but I saw green, so I know it must
have been an H."
We originally tested two synesthetes, and found that they performed better
than non-synesthetes on these two tasks. However, further testing with a
larger sample showed that only some of the synesthetes performed better
than non-synesthetes. Based on these results (here's where the numcog list
really comes in) we proposed that synesthesia arises from a genetically
mediated failure of pruning between adjacent brain regions involved in
recognizing letters and numbers in the fusiform gyrus, and adjacent regions
involved in the processing of colors (V4/V8/hV4, there's a huge controversy
about this in the color vision literature).
However, since this time, we have continued to interview and test
additional synesthetes, and we find that, while some of them do show the
improved behavioral performance we found in our first few synesthetes, not
all of them do. Interestingly, when we went back and looked more carefully
at their experiential reports, we found that those that performed better
than controls reported that they experienced colors for letters and
numbers, that they experienced these colors out in the world, and that they
experienced colors for only these classes of stimuli. However, in the
synesthetes that did not perform better than controls, we found that they
reported that their colors were seen in the minds eye, were often
experienced for days of the week and months of the year in addition to
numbers, but often did not experience colors for letters, and reported
experiencing spatial forms for the numbers and calendars (for example,
numbers might be visualized as an ascending staircase, going up and to the
left, or the months of the year might be experienced as having a racetrack
shape, with each month colored differently).
Based on the different stages of numerical processing in the triple-code
model, we proposed that these "higher" synesthetes perhaps had
cross-activation in the HIPS/angular gyrus region, instead of in the
fusiform region we had proposed for the "lower" synesthetes that we tested
first, and that it is this cross-activation between higher, more abstract,
stages of numerical processing, and spatial representations in the PSPL
(see, for example, Anna's work on this topic) that leads to the spatial
experiences when thinking of letters and numbers that these synesthetes
report.
Current fMRI research, conducted in collaboration with Geoff Boynton at the
Salk Institute supports this general distinction. We find that, for all
synesthetes we have tested, activity in color selective regions is greater
when viewing graphemes than viewing non-linguistic stimuli matched for
visual complexity (Mauro Pesenti kindly lent us his stimuli). No such
difference was observed for control subjects. Additionally, we find that
there is a great deal of variability in the responses in V1 for different
synesthetes, and that the fMRI response in early visual areas (V1-V4)
predicts behavioral performance on our psychophysical tasks. Converging
ERP data on a subset of the same synesthetes by Noam Sagiv and Lynn
Robertson at UC Berkeley also suggests that the differences in behavioral
performance and self-report map onto differences in the neural locus of
synesthesia.
Because of these results, I have become more and more interested in the
connection between spatial representation and numerical representation, and
will be beginning my post-doc in the fall of 2004 with Stan Dehaene
examining the relationship between spatial processing and numerical
processing in the parietal lobe through the use of fMRI. In addition, I
will continue to work with synesthesia, although less intensely, to try to
make more sense of the spatial forms observed in certain forms of
synesthesia.
Sorry to have rambled on so long, but synesthesia still isn't that well
known, even among cognitive neuroscientists, and I wanted to give a bit
more background on this work so that people on the list will understand
where I am coming from.
Cheers,
Ed
Edward M. Hubbard, MA
Brain and Perception Laboratory
University of California, San Diego
9500 Gilman Dr. 0109
La Jolla, CA 92093-0109
and
SNL-B
Salk Institute for Biological Studies
10010 North Torrey Pines Road
La Jolla, California 92037-1099
edhubbard@xxxxxxxxxxxx
http://psy.ucsd.edu/~edhubbard
-------------------------------------------------------
This message has been brought to you by the numcog mailing
list at freelists.org. If you would like to unsubscribe from
the list, send an email to numcog-request@xxxxxxxxxxxxx and put
"unsubscribe" (without the quotes) in the Subject line of the email.
A digest version is available, to sign up login to your account at
//www.freelists.org/login.html, and see instructions in the FAQ or
welcome files.
For other administrative queries, email Anna:
ajwilson@xxxxxxxxxxxxxxxxxxxx
Other related posts:
