Computers and Education Artificial Intelligence,
Год журнала:
2024,
Номер
7, С. 100266 - 100266
Опубликована: Июль 17, 2024
This
systematic
review
aims
to
explore
published
research
on
the
use
of
ChatGPT
in
language
learning
between
November
2022
and
2023,
outlining
types
papers,
methodologies
adopted,
publishing
journals,
major
trends,
topics
interest,
existing
gaps
demanding
attention.
The
PRISMA
framework
was
utilized
capture
latest
articles,
selecting
36
articles
that
met
inclusion
criteria.
Findings
extracted
from
this
include
(1)
authors
worldwide
contribute
topic,
with
Asia
North
America
leading;
(2)
wide
distribution
across
various
journals
underscores
interdisciplinary
nature
such
as
computer
science,
psychology,
linguistics,
education,
other
social
sciences;
(3)
empirical
dominates
literature
is
published,
majority
focusing
higher
education
ethical
considerations.
Other
findings
plays
multifaceted
roles,
supporting
self-directed
learning,
content
generation,
teacher
workflows.
Research
need
for
diversified
scopes,
longitudinal
studies,
exploration
stakeholders'
perceptions,
assessments
feedback
quality.
Computers and Education Artificial Intelligence,
Год журнала:
2023,
Номер
5, С. 100172 - 100172
Опубликована: Янв. 1, 2023
The
accessibility
of
advanced
Artificial
Intelligence-based
tools,
like
ChatGPT,
has
made
Large
Language
Models
(LLMs)
readily
available
to
students.
These
LLMs
can
generate
original
written
content
assist
students
in
their
academic
assessments.
With
the
rapid
adoption
LLMs,
exemplified
by
popularity
OpenAI's
there
is
a
growing
need
explore
application
education.
Few
studies
examine
students'
use
as
learning
tools.
This
paper
focuses
on
ChatGPT
engineering
higher
education
through
an
in-depth
case
study.
It
investigates
whether
high-quality
university
essays
with
assistance,
existing
identification
systems
detect
produced
and
how
perceive
usefulness
acceptance
learning.
research
adopts
deductive/inductive
approach,
combining
conceptualization
empirical
evidence
analysis.
study
involves
mechanical
management
students,
who
compose
using
LLMs.
essay
assessment
showed
good
results,
but
some
recommendations
emerged
for
teachers
Thirteen
detectors
were
tested
without
achieving
satisfactory
suggesting
avoid
ban.
In
addition,
administered
questionnaire
based
constructs
items
that
follow
technology
models
literature.
results
contribute
qualitative
highlighting
possible
future
educational
practices.
Frontiers in Education,
Год журнала:
2024,
Номер
9
Опубликована: Фев. 9, 2024
Introduction
The
integration
of
ChatGPT,
an
advanced
AI-powered
chatbot,
into
educational
settings,
has
caused
mixed
reactions
among
educators.
Therefore,
we
conducted
a
systematic
review
to
explore
the
strengths
and
weaknesses
using
ChatGPT
discuss
opportunities
threats
in
teaching
learning.
Methods
Following
PRISMA
flowchart
guidelines,
51
articles
were
selected
819
studies
collected
from
Scopus,
ERIC
Google
Scholar
databases
period
2022-2023.
Results
synthesis
data
extracted
included
revealed
32
topics
including
13
strengths,
10
weaknesses,
5
4
We
used
Biggs’s
Presage-Process-Product
(3P)
model
learning
categorize
three
components
3P
model.
Discussion
In
Presage
stage,
analyzed
how
interacts
with
student
characteristics
contexts
ensure
that
technology
adapts
effectively
diverse
needs
backgrounds.
Process
impacted
activities
determine
its
ability
provide
personalized,
adaptive,
effective
instructional
support.
Finally,
Product
evaluated
contributed
outcomes.
By
carefully
considering
application
each
stage
learning,
educators
can
make
informed
decisions,
leveraging
addressing
optimize
processes.
International Journal of Educational Technology in Higher Education,
Год журнала:
2024,
Номер
21(1)
Опубликована: Март 4, 2024
Abstract
In
the
evolving
landscape
of
higher
education,
challenges
such
as
COVID-19
pandemic
have
underscored
necessity
for
innovative
teaching
methodologies.
These
catalyzed
integration
technology
into
particularly
in
blended
learning
environments,
to
bolster
self-regulated
(SRL)
and
higher-order
thinking
skills
(HOTS).
However,
increased
autonomy
can
lead
disruptions
if
issues
are
not
promptly
addressed.
this
context,
OpenAI's
ChatGPT,
known
its
extensive
knowledge
base
immediate
feedback
capability,
emerges
a
significant
educational
resource.
Nonetheless,
there
concerns
that
students
might
become
excessively
dependent
on
tools,
potentially
hindering
their
development
HOTS.
To
address
these
concerns,
study
introduces
Guidance-based
ChatGPT-assisted
Learning
Aid
(GCLA).
This
approach
modifies
use
ChatGPT
settings
by
encouraging
attempt
problem-solving
independently
before
seeking
assistance.
When
engaged,
GCLA
provides
guidance
through
hints
rather
than
direct
answers,
fostering
an
environment
conducive
SRL
A
randomized
controlled
trial
(RCT)
was
employed
examine
impact
compared
traditional
foundational
chemistry
course
within
setting.
involved
61
undergraduate
from
university
Taiwan.
The
findings
reveal
enhances
SRL,
HOTS,
construction
use.
results
directly
align
with
research
objective
improve
outcomes
providing
answers
ChatGPT.
conclusion,
introduction
has
only
facilitated
more
effective
experiences
environments
but
also
ensured
engage
actively
journey.
implications
highlight
potential
ChatGPT-based
tools
enhancing
quality
essential
self-regulation
Furthermore,
offers
insights
regarding
education.
IEEE Transactions on Learning Technologies,
Год журнала:
2024,
Номер
17, С. 1683 - 1700
Опубликована: Янв. 1, 2024
While
ongoing
efforts
have
continuously
emphasized
the
integration
of
ChatGPT
with
science
teaching
and
learning,
there
are
limited
empirical
studies
exploring
its
actual
utility
in
classroom.
This
study
aims
to
fill
this
gap
by
analyzing
lesson
plans
developed
29
pre-service
elementary
teachers
assessing
how
they
integrated
into
learning
activities.
We
first
examined
was
subject
domains,
methods/strategies
then
evaluated
using
a
GenAI-TPACK-based
rubric.
further
teachers'
perceptions
concerns
about
integrating
learning.
Results
show
diverse
number
applications
different
domains—e.g.,
Biology
(9/29),
Chemistry
(7/29),
Earth
Science
(7/29).
Fourteen
types
were
identified
plans.
On
average,
scored
high
on
modified
TPACK-based
rubric
(M
=
3.29;
SD
.91;
1-4
scale),
indicating
reasonable
envisage
particularly
'instructional
strategies
&
ChatGPT'
3.48;
.99).
However,
relatively
lower
exploiting
ChatGPT's
functions
toward
full
potential
3.00;
.93),
compared
other
aspects.
also
several
inappropriate
use
cases
planning
(e.g.,
as
source
hallucinated
internet
material
technically
unsupported
visual
guidance).
Pre-service
anticipated
afford
high-quality
questioning,
self-directed
individualized
support,
formative
assessment.
Meanwhile,
expressed
accuracy
risks
that
students
may
be
overly
dependent
ChatGPT.
They
suggested
solutions
systemizing
classroom
dynamics
between
students.
The
underscores
need
for
more
research
roles
generative
AI
settings
provides
insights
future
AI-integrated
Asia Pacific Journal of Education,
Год журнала:
2024,
Номер
44(1), С. 139 - 155
Опубликована: Янв. 2, 2024
Generative
artificial
intelligence
has
been
regarded
as
a
transformative
tool.
While
responsible
and
ethical
applications
could
bring
opportunities
to
education,
their
misuse
pose
demanding
challenges.
It
is
necessary
clarify
the
technological
affordances
challenges
in
normative
way
lay
foundation
for
future
development.
This
study
addressed
dearth
of
literature
by
performing
systematic
review,
aiming
(i)
explore
utility
availability
from
perspective;
(ii)
summarize
current
risks
prevention;
(iii)
propose
possible
directions
research
practice.
A
total
27
academic
articles
published
core
journals
between
2020
2023
were
analyzed,
inductive
grounded
approach
was
used
categorize
coding
schemes.
The
findings
revealed
four
affordances:
accessibility,
personalization,
automation,
interactivity;
five
challenges:
integrity
risk,
response
errors
bias,
over-dependence
widening
digital
divide,
privacy
security.
We
directions,
encourage
educational
organizations
formulate
guidelines
use
AI
call
on
educators
embrace
trends
education
instead
shunning
its
use,
guide
students
treat
it
thought
aid
reference,
rather
than
relying
entirely.
Abstract
Over
the
last
four
decades,
studies
have
investigated
incorporation
of
Artificial
Intelligence
(AI)
into
education.
A
recent
prominent
AI-powered
technology
that
has
impacted
education
sector
is
ChatGPT.
This
article
provides
a
systematic
review
14
empirical
incorporating
ChatGPT
various
educational
settings,
published
in
2022
and
before
10th
April
2023—the
date
conducting
search
process.
It
carefully
followed
essential
steps
outlined
Preferred
Reporting
Items
for
Systematic
Reviews
Meta-Analyses
(PRISMA
2020)
guidelines,
as
well
Okoli’s
(Okoli
Commun
Assoc
Inf
Syst,
2015)
rigorous
transparent
review.
In
this
review,
we
aimed
to
explore
how
students
teachers
utilized
primary
findings
those
studies.
By
employing
Creswell’s
(Creswell
Educational
research:
planning,
conducting,
evaluating
quantitative
qualitative
research
[Ebook],
Pearson
Education,
London,
coding
techniques
data
extraction
interpretation,
sought
gain
insight
their
initial
attempts
at
approach
also
enabled
us
extract
insights
considerations
can
facilitate
its
effective
responsible
use
future
contexts.
The
results
show
learners
virtual
intelligent
assistant,
where
it
offered
instant
feedback,
on-demand
answers,
explanations
complex
topics.
Additionally,
used
enhance
writing
language
skills
by
generating
ideas,
composing
essays,
summarizing,
translating,
paraphrasing
texts,
or
checking
grammar.
Moreover,
turned
an
aiding
tool
directed
personalized
learning
assisting
understanding
concepts
homework,
providing
structured
plans,
clarifying
assignments
tasks.
However,
specific
(n
=
3,
21.4%)
overuse
may
negatively
impact
innovative
capacities
collaborative
competencies
among
learners.
Educators,
on
other
hand,
create
lesson
generate
quizzes,
provide
additional
resources,
which
helped
them
productivity
efficiency
promote
different
teaching
methodologies.
Despite
these
benefits,
majority
reviewed
recommend
importance
training,
support,
clear
guidelines
both
educators
mitigate
drawbacks.
includes
developing
critical
evaluation
assess
accuracy
relevance
information
provided
ChatGPT,
strategies
integrating
human
interaction
collaboration
activities
involve
AI
tools.
Furthermore,
they
ongoing
proactive
dialogue
with
policymakers,
stakeholders,
practitioners
refine
environments.
could
serve
insightful
resource
who
seek
integrate
stimulate
further
field.
IEEE Transactions on Learning Technologies,
Год журнала:
2024,
Номер
17, С. 1279 - 1289
Опубликована: Янв. 1, 2024
Generative
AI
has
emerged
as
a
noteworthy
milestone
and
consequential
advancement
in
the
annals
of
major
disciplines
within
domains
human
science
technology.
This
study
aims
to
explore
effects
generative
AI-assisted
pre-service
teaching
skills
training
on
teachers'
self-efficacy
higher
order
thinking.
The
participants
this
were
215
mathematics,
science,
computer
teachers
from
university
China.
Firstly,
pretest-posttest
quasi-experimental
design
was
implemented
for
an
experimental
group
(teaching
by
AI)
control
traditional
methods)
investigating
teacher
thinking
two
groups
before
after
experiment.
Secondly,
semi-structured
interview
comprising
open-ended
questions
administered
25
present
their
views
teaching.
results
showed
that
scores
group,
who
used
professional
development,
considerably
than
those
both
(F
=
8.589,
p
0.0084<
0.05)
7.217,
0.008
<
0.05).
It
revealed
can
be
effective
supporting
development.
produced
practical
development
method
with
AI.
Journal of Educational Computing Research,
Год журнала:
2024,
Номер
62(6), С. 1509 - 1537
Опубликована: Апрель 17, 2024
The
field
of
computer-assisted
language
learning
has
recently
brought
about
a
notable
change
in
English
as
Foreign
Language
(EFL)
writing.
Starting
from
October
2022,
students
across
different
academic
fields
have
increasingly
depended
on
ChatGPT-4
helpful
resource
for
addressing
particular
challenges
EFL
This
study
aimed
to
investigate
the
use
and
acceptance
students’
To
this
end,
an
experiment
was
conducted
with
76
undergraduate
private
school
Algeria.
participants
were
randomly
allocated
into
two
groups:
experimental
group
(n
=
37)
control
39).
Additionally,
questionnaire
administered.
results
showed
that
(EG)
outperformed
(CG).
Besides,
findings
revealed
EG
post-test
their
pre-test
scores.
also
substantial
improvements
EG’s
views
perceived
usefulness,
ease
use,
attitudes,
behavioral
intention.
According
results,
helped
boost
students'
writing
skills,
which
ultimately
led
acceptance.
Students
appear
particularly
interested
because
its
potential
usefulness
putting
what
they
learn
practice.
Some
suggestions
recommendations
provided.