Joy Geren
Grammatical Development in Children with Cochlear Implants
Past research has found that deaf children with cochlear implants show a
grammatical deficit that exceeds their semantic deficit. This study explores
the nature of the grammatical deficit by examining morphological and syntactic
abilities in 3- to 9-year-old children with implants. By comparing performance
across areas we hope to learn more about the process of language learning in
this unique population.
|
|
|
Steve Franconeri & George Alvarez
The Magic Number 4 in Visual Attention
Performance across several visual tasks, such as memory for spatial
position,multiple object tracking, or rapid counting (subitizing), often seems to
suggest that the visual system can localize or track a fixed number of objects
at once. These fixed capacities, usually of about 3 or 4 objects, are often
taken as strong constraints on possible architectures of visual attention.
However, for each of these tasks, there is debate over whether capacities are
truly fixed. We critically examine these limits, and show that manipulations of
task difficulty can push these limits down to 1 object, or up to 8 objects. We
discuss the possible architectural roots of this limit, as well as potentially
related limits in the binding of object features.
|
|
|
Rachel Garoff
Remembering Items That Were Never Encountered: The Neural Basis of False
Memory
False recognition occurs when one mistakenly claims to remember something
that has never been previously encountered. However, several distinct
types of false recognition exist. For instance, gist-based false
recognition occurs when one claims to remember something that is related
to previously encountered items, whereas baseline false recognition occurs
when one claims to remember something that is unrelated to anything that
was previously encountered. While recent neuroimaging work has begun to
explore the neural basis of memory distortion, research has yet to
disentangle the neural correlates differentially associated with these two
types of memory errors. To address this issue, I will first discuss an
fMRI experiment that directly compared the neural substrates associated
with the retrieval of gist-based false memories and baseline false
memories. Then I will turn to the creation of false memories. A major
difference between these two types of memory errors is that gist-based
false recognition can be traced back to similar events that occurred
during memory encoding, while baseline false recognition is not directly
associated with any specific encoding events. To further understand this
link between encoding events and subsequent gist-based false recognition,
I will discuss an fMRI experiment that examined the encoding-related
neural activity associated with subsequent gist-based false recognition.
|
|
|
Won Mok Shim
The Role of Attention in Position Encoding
Contrary to both our subjective experience of a stable world and established knowledge about the retinotopic maps in
the visual areas of the brain, recent findings show that perceived spatial position can be distorted by motion perception and
eye movements. In this talk, I will propose that visual attention, critically involved in both motion perception and eye
movements, may be a common mechanism underlying both motion-induced and eye movement-induced position distortions. To address
this, I will present three sets of experiments examining the influence of attention on the organization of the visual
representation of space in neighboring areas. I will first show that the high-level percept of motion can modulate the
position shift, even if the low-level spatio-temporal properties of the motion stimulus remain constant, by using an ambiguous
motion stimulus. Then I will demonstrate the role of attention more directly using the attentive tracking paradigm which
allows explicit control over the locus of attention. Finally, I will discuss the role of attention in peri-saccadic spatial
compression, in which a spatial structure presented around the time of an eye movement is compressed so that it appears to
have been presented closer to the eye movement target. The results of the experiments replicate the compression effect using
only a shift of attention without eye movements suggesting that compression may be driven by attention that precedes the eyes
to the target.
|
|
|
Helen Tager-Flusberg
Language in Autism
In this presentation I will review the characteristics of language disorder
in autism that are (a) universal and specific to autism spectrum disorders;
and (b) more variable aspects of language impairment that do not define the
syndrome. Universal impairments are found in pragmatics and communication.
These are closely linked to social deficits and problems in understanding
other minds. Structural language impairments define different subgroups
within ASD. One subgroup has impairments that parallel those found in
specific language impairment (SLI). The key features of this subgroup will
be described including evidence from behavioral and neuroimaging studies.
Implications for genetic studies of autism, SLI and language will be
discussed.
|
|
|
Gary Marcus
What is the Language Faculty Made Of?
Evidence from Human Infants and Molecular Biology
To what extent is the machinery for acquiring language special to
language? Discussion has typically teetered between two extreme views, the
notion that language is an autonomous specialized module, and the view
that language is just a special application of sophisticated cognitive
processes. In this talk, I argue against both of these views and in favor
of a new perspective, dubbed the "Orchestrated Components" view. According
to the Orchestrated Components view, language is to be understood as a
coordinated product of a diverse set of mechanisms, some evolutionarily
ancient, some evolutionarily more recent.
The first part of the talk motivates this view, drawing on recent
literature in genetics and developmental neuroscience. In the second part,
I present results from empirical investigations of the learning
capacities of human infants, suggesting that some learning mechanisms are
domain-general, while others are (at least initially) domain-specific.
Taken together, these empirical and theoretical considerations argue for
an approach to linguistics that makes closer contact with cognition, yet
does not abandon the notion that the capacity for language arises from a
human-specific endowment.
|
|
|
Dave Barner
Pluralities: Set Representations in Natural Language and Early Cognition
I will present research on children's early understanding of set
representations and how they interface with object-based attention and
singular-plural morphology in language. Nouns that are subject to a
singular-plural distinction (i.e. count nouns) specify number as a measuring
dimension, but do not represent magnitude in the way of pre-linguistic
number systems. As a result, they offer a novel format for number
representation in cognitive development.
The special case of set representation by count nouns will be contrasted
with nouns that do not specify a uniform measuring dimension (i.e. mass
nouns). Quantity judgment data from 3 and 4-year-old children and adults
will be used to make the case that the distinction between count and mass
nouns is asymmetric from early in language acquisition. A comparison of
quantity judgments for speakers of French, English and Japanese will further
suggest that a single structure of grammar licenses individuation in
languages that have count nouns. Finally, data from English-speaking
children will suggest a mechanism for the diachronic origins of
cross-linguistic differences in how nouns are used.
|
|
|
Mathieu LeCorre
On the Integration of Numerals and Core Knowledge of Number in the Construction of Arithmetic Competence
The adult cognitive capacity for arithmetic is founded upon at least two representational systems. One of these, analog magnitudes, is
innate and shared with many other animals. The other, the verbal count list, is a cultural construction acquired in early childhood. In
adulthood, linguistic symbols for the integers (verbal numerals and Arabic digits) are mapped onto analog magnitudes (AMs). This mapping
is an important part of adult arithmetic competence and its acquisition an important part of its development. I explored the development
of this mapping in two ways. First, I asked whether children must create a mapping between numerals to learn how to use counting to
represent number. In particular, since children know an ordered count list long before they know how to use it to determine the number
of objects in a set, I asked whether children map the numerals in their count list onto AMs before they learn how counting represents
number. Drawing upon evidence from multiple tasks designed to assess the development of the mapping of numerals to magnitudes, I will
argue that children do not map numerals onto AMs to learn how to use verbal counting as a representation of number.
Having shown that the development of the mapping does not play a role in the acquisition of counting, I will explore whether it
plays a role in the development of children's knowledge of the numerical order of the numerals in the count list (i.e. that "ten fish"
is more than "six fish"). I will present evidence that learning how counting represents number isn't sufficient to know how numerals are
ordered numerically (e.g. "six fish" is less than "ten fish"), and that it's not until children have mapped some numerals onto analog
magnitudes that they can make ordinal inferences (e.g. determine that a box with "ten fish" has more fish in it than a box with "six
fish") from verbal descriptions of the numerosity of sets.
|
|
|
Jodi Davenport
Seeing Scenes: Context Effects on Object and Background Perception
In the world objects and settings tend to co-occur. Cars are usually found
on roads, not in kitchens. Does knowlege about which objects and settings
usually appear together influence the initial perception of a briefly
presented scene?
In two series of studies, color photographs were used to investigate the
influence of the semantic consistency between objects and settings on
perception. Objects were reported more accurately when they appeared in
typical rather than unusual settings. Additionally, settings were reported
more accurately when they contained plausible foreground objects. Objects
and scenes appear to be processed interactively, not in isolation.
|
|
|
Ben Balas
Texture Synthesis and Perception: Using
Computational Models to Study Texture Representations in the Human
Visual System
Traditionally, texture perception has been studied using
artificial textures made of random dots or repeated shapes. At the same
time, computer algorithms for natural texture synthesis have improved
dramatically. We suggest that such models of texture synthesis provide a
useful means for studying the perception of real textures. We demonstrate
the utility of this approach through a psychophysical assessment of a
particular computational model, providing insight into which statistics
are most vital for natural texture perception. We employ Portilla and
Simoncelli's texture synthesis algorithm for this purpose, which is a
parametric model that mimics computations carried out in the human visual
system. We find an intriguing interaction between texture type (periodic
v. structured) and image statistics (autocorrelation function and filter
magnitude correlations), suggesting different representations may be
employed for these texture families under pre-attentive viewing.
|
|
|
Richard Russell
The Importance of Pigmentation for Face Perception
In the object recognition literature, there is an explicit belief, based on
experimental evidence, that pigmentation cues are relatively
unimportant. Though the same belief has not been articulated explicitly
with respect to faces, there is evidence of a widespread implicit
assumption that pigmentation is unimportant for face perception, despite a
lack of supporting experimental evidence. I will describe a systematic
investigation of the utility of pigmentation cues in three judgments about
faces--recognition, sex classification, and attractiveness--and will
present evidence that pigmentation cues do in fact play an important role
in each of these judgments. Interspersed thoughout the investigation are
results relating pigmentation to other phenomena, including the effects of
inversion and negation on face recognition, and the use of cosmetics.
|
|
|
Peter Gordon
Why Issaoui Can't Count: Linguistic or Cultural Causes of Innumeracy in the Pirahã
This talk will describe a set of studies examining numerical abilities
in the Pirahã, a tribe living in lowland Amazonia in Brazil. Being one
of the least integrated of the indigenous tribes in modern times, the
Pirahã still live a basic hunter-gatherer existence with little or no
contact with mainstream Brazilian culture. Their language contains no
words that indicate exact numerosity with only two words, ho'i and hoi'
that indicate approximately one and two respectively. The studies
reported in this talk examine the question of whether such lack of words
for exact number leads to "strong determinism" in the sense that the
speakers of this language are unable to entertain exact numerical
concepts. The pattern of results suggests that the Pirahã inherit only
the innate abilities for exact appreciation of numbers smaller than 3,
and approximate appreciation of larger numbers. The failure of both the
language and conceptual structures to encode exact numerosity suggests
that languages can be incomplete and incommensurable --contrary to the
assumptions of Whorf, Sapir and Boas. I discuss whether the pattern of
results show a true effect of linguistic determinism as opposed to revealing
holistic cultural factors that conspire to limit the range of general cognition
in the Pirahã tribe.
|
|
