Integrative modeling enables ChatGPT to achieve average level of human counselors performance in mental health Q&A
Information Processing & Management,
Год журнала:
2025,
Номер
62(5), С. 104152 - 104152
Опубликована: Апрель 6, 2025
Язык: Английский
Screening Social Anxiety with the Social Artificial Intelligence Picture System
Journal of Anxiety Disorders,
Год журнала:
2024,
Номер
109, С. 102955 - 102955
Опубликована: Дек. 6, 2024
Язык: Английский
A psychologically interpretable artificial intelligence framework for the screening of loneliness, depression, and anxiety
Applied Psychology Health and Well-Being,
Год журнала:
2024,
Номер
17(1)
Опубликована: Дек. 19, 2024
Negative
emotions
such
as
loneliness,
depression,
and
anxiety
(LDA)
are
prevalent
pose
significant
challenges
to
emotional
well-being.
Traditional
methods
of
assessing
LDA,
reliant
on
questionnaires,
often
face
limitations
because
participants'
inability
or
potential
bias.
This
study
introduces
emoLDAnet,
an
artificial
intelligence
(AI)-driven
psychological
framework
that
leverages
video-recorded
conversations
detect
negative
through
the
analysis
facial
expressions
physiological
signals.
We
recruited
50
participants
undergo
questionnaires
interviews,
with
their
responses
recorded
video.
The
emoLDAnet
employs
a
combination
deep
learning
(e.g.,
VGG11)
machine
decision
trees
[DTs])
identify
states.
incorporates
OCC-PAD-LDA
transformation
model,
enhancing
interpretability
AI
decisions
by
translating
into
psychologically
meaningful
data.
Results
indicate
achieves
high
detection
rates
for
anxiety,
F1-scores
exceeding
80%
Kendall's
correlation
coefficients
above
0.5,
demonstrating
strong
agreement
traditional
scales.
underscores
importance
model
in
improving
screening
accuracy
impact
classifiers
framework's
performance.
has
support
large-scale
well-being
early
contribute
advancement
mental
health
care.
Язык: Английский
A new era for stress research: supporting user performance and experience in the digital age
Ergonomics,
Год журнала:
2024,
Номер
unknown, С. 1 - 34
Опубликована: Ноя. 8, 2024
Stress
is
both
a
driver
of
objective
performance
impairments
and
source
negative
user
experience
technology.
This
review
addresses
future
directions
for
research
on
stress
ergonomics
in
the
digital
age.
The
structured
around
three
levels
analysis.
At
individual
level,
elicited
by
novel
technologies
tasks
including
interaction
with
AI
robots,
working
Virtual
Reality,
operating
autonomous
vehicles.
organisational
novel,
potentially
stressful
challenges
include
maintaining
cybersecurity,
surveillance
monitoring
employees
supported
technology,
addressing
bias
discrimination
workplace.
sociocultural
values
norms
are
evolving
symbiotically,
raising
demands
illustrated
respect
to
interactions
social
media
new
ethical
challenges.
We
also
briefly
promise
neuroergonomics
emotional
design
support
mitigation.
conclude
seven
high-level
principles
that
may
guide
work.
Язык: Английский