文件4 Interactive ChatGPT Example

Interactive ChatGPT Example
給 ChatGPT 的 prompt
Design a grading rubric for the following assignment, which is for a 3rd year
university course on Judgement & Decision Making. The rubric must include the
following RUBRIC COMPONENTS and address the LEARNING
OBJECTIVES for the course. Be sure to use UK English.
ASSIGNMENT DETAILS:
ChatGPT Assignment 1
Generating and Evaluating Examples of Concept Transfer
Objective:
To analyse AI-generated examples, assess their accuracy in demonstrating a

concept from the first half of the course, and enhance the transfer of knowledge
by deepening your understanding through critical thinking and application.
Concept:
Choose any concept from the first half of the course (excluding sunk cost). Classes
from the first half of the course include: introduction to artificial intelligence and
its relationship to cognition, heuristics and biases, thinking fast and slow, non-
human feats of intelligence, reasoning and problem-solving, and insight and
metacognition.
Primary Goal:
Transfer of knowledge on the chosen concept to various contexts, enabling

students to apply the concept effectively outside the classroom.
Instructions:
1. Choose a concept from the first half of the course and generate AI-
generated examples of the chosen concept in 3 different scenes using
ChatGPT or Bing Chat. To do this, enter a prompt asking the AI to
demonstrate the chosen concept in 3 different scenes or situations.
o Prompts:a. Demonstrate the concept of sunk cost in the form of a
script for a podcast where two hosts discuss examples of sunk cost in
their lives.b. Demonstrate the concept of sunk cost in the form of a
dialogue for a superhero movie where a character faces the
consequences of their sunk cost.c. Demonstrate the concept of sunk
cost in the form of an email to your boss where you propose to
abandon a project that has become a sunk cost.
o Here are a few more scenes using the sunk cost concept, which you
can use for inspiration. But note that the topic of sunk cost is off the
table.
▪ a song lyric about sunk cost
▪ a monologue for a comedy show where a character makes fun
of their own sunk cost
▪ a dialogue for a drama series where a character confronts their
partner about their sunk cost
▪ an article for a newspaper where you analyse the sunk cost of a
political decision
▪ a script for a documentary where you interview people who
have experienced sunk cost
▪ an email to your boss where you propose to abandon a project
that has become a sunk cost
▪ a letter to your future self where you warn them about the
dangers of sunk cost
▪ a dialogue for a sci-fi movie where a character travels back in
time to change their sunk cost
2. After receiving the AI-generated examples, read them carefully and assess
their accuracy in demonstrating the chosen concept. Make sure to save the
transcripts of your interactions with ChatGPT or Bing Chat.
3. Rank the AI-generated examples from most accurate, somewhat accurate, to
least accurate based on their effectiveness in demonstrating the chosen
concept.
4. Write an 800-word analysis explaining your ranking and evaluating the
examples, following these steps:
a. Justify your ranking by explaining why you ranked the examples in the
order you did.
b. Focus on the least accurate example, and answer the following questions:
o Which aspects of the chosen concept did the AI capture correctly in

this scenario?
o Suggest improvements to the example to better demonstrate the
concept.
o Identify the aspects of the chosen concept that the AI failed to
capture in this scenario.c. If, instead, you find all 3 examples to be
accurate, choose 1 of them and explain how it effectively
demonstrates every aspect of the chosen concept. In this case, you
don’t need to complete step 4b.
5. Write a 500-word reflection on how the chosen concept can be applied to
other contexts in your life or future experiences. Consider how analysing the
AI-generated examples has helped you develop a deeper understanding of
the concept and its applications.
Important:
Do not use ChatGPT, Bing Chat, or any large language models to help you write
the analysis or reflection. Use it only for generating examples and clarifying
concepts if needed.
RUBRIC COMPONENTS:
1. Task description or descriptive title:
Begin with a clear description of the task or assignment that students are
expected to produce or perform. This sets the context for the rubric and
informs students about the aspects that will be assessed and how they will
be assessed.
2. Criteria or dimensions:
Identify specific criteria that reflect the characteristics or components of the

task to be rated. These criteria should encompass the skills, knowledge, and
behaviours that students are expected to demonstrate throughout the
assignment or project.
3. Levels of achievement or mastery scale:
Establish a scale that defines the levels of achievement or mastery for each
criterion, such as exceeds expectations, meets expectations, and does not
meet expectations. Include a score allocation for each level to quantify
student performance.
4. Performance descriptors:
Provide descriptions of the performance quality for each criterion at each

level of achievement. These descriptors should give students a clear idea of
what they must do to demonstrate a certain level of mastery, understanding,
or proficiency in each area.
5. Weighting of criteria:
Assign appropriate weights to each criterion according to its significance
relative to the overall learning objectives. This ensures that the total score
accurately reflects the student's performance in all aspects of the task.
LEARNING OBJECTIVES:
1. Gain a thorough understanding of the fundamental concepts and theories in

the field of judgement and decision making, including biases and heuristics,
dual-process theories, non-human “feats of intelligence,” reasoning and
problem solving, insight and metacognition, and decision making under
uncertainty.
2. Analyse and apply these theories to real-world situations, such as high-stakes
decision making in fields like medicine, law, and education, using large
language models like ChatGPT to improve decision-making abilities and
critically evaluate decision-making processes.
3. Enhance their ability to identify and avoid common biases and heuristics in
their own decision-making by working with AI chatbots and strategies
developed in ChatGPT assignments, and develop and implement effective
decision-making strategies.
4. Evaluate and analyse decision-making strategies, including group decision
making and nudging, and understand the impact of brain training, transfer,
and mental representations on expertise, using critical thinking and feedback
skills developed in ChatGPT assignments.
5. Effectively communicate their understanding of judgement and decision
making to others, using the communication skills developed in the
ChatGPT assignments and the flashcard creation assignment, and
persuading them to adopt new approaches to decision making.
6. Understand the research methods that are used to study judgement and
decision-making processes in the laboratory and in real-world settings, and
critically evaluate research studies related to the field.
7. Assess, use, document, and present information objectively and effectively,
including synthesising information from multiple sources and presenting
findings in a clear and concise manner.
8. Demonstrate the ability to argue for a position in the debate about cognitive
processes and the methods used to investigate them, using critical thinking
and persuasive communication skills developed throughout the course.
ChatGPT 的回應
Rubric for Assignment 1: Generating and Evaluating Examples of Concept

Transfer
Component 1: Understanding of Concept (25%)
Criteria:
• Clear understanding of the chosen concept demonstrated through the
explanation and application in different scenarios.
Levels of Achievement:
Score Description
Chosen concept is clearly defined and explained.

Exceeds Excellent understanding is demonstrated
Expectations (21- through the generation of accurate, diverse, and
25 pts) complex AI-generated examples.
Chosen concept is adequately defined and

Meets explained. Good understanding is demonstrated
Expectations (16- through the generation of mostly accurate AI-
20 pts) generated examples.
Chosen concept is poorly defined and explained.

Does Not Meet Limited understanding is demonstrated through
Expectations (0- the generation of inaccurate or overly simplistic
15 pts) AI-generated examples.
Component 2: Evaluation of AI-Generated Examples (25%)
Criteria:
• Demonstrated ability to critically analyse and evaluate the AI-generated
examples.
Score Description
Exceeds Detailed, insightful analysis provided for each AI-

Expectations generated example. Justifications for rankings are
(21-25 pts) sound and reflect deep understanding of the
Score Description
concept. Proposed improvements are highly

effective in better demonstrating the concept.
Adequate analysis provided for each AI-generated

example. Justifications for rankings are
Meets reasonable. Proposed improvements are
Expectations somewhat effective in better demonstrating the
(16-20 pts) concept.
Limited or no analysis provided for AI-generated

Does Not Meet examples. Justifications for rankings lack clarity
Expectations (0- or logic. Proposed improvements are ineffective
15 pts) or not provided.
Component 3: Application of Concept to Other Contexts (25%)
Criteria:
• Demonstrated ability to transfer knowledge and apply the chosen concept
to other contexts.
Score Description
Exceptional demonstration of knowledge

Exceeds transfer, with detailed examples of how the
Expectations (21- chosen concept can be applied to other contexts
25 pts) in the student's life or future experiences.
Good demonstration of knowledge transfer, with

Meets some examples of how the chosen concept can
Expectations (16- be applied to other contexts in the student's life
20 pts) or future experiences.
Limited or no demonstration of knowledge

Does Not Meet transfer. Few or no examples of how the chosen
Expectations (0- concept can be applied to other contexts in the
15 pts) student's life or future experiences.
Component 4: Clarity and Quality of Writing (25%)

Criteria:
• Clarity and precision of language.
• Proper citation and use of sources.
• Overall organisation and coherence of the assignment.
Score Description
Assignment is exceptionally clear, well-

Exceeds structured, and well-written, with no
Expectations (21- grammatical or stylistic errors. Citations and
25 pts) sources are properly used and documented.
Assignment is mostly clear and well-structured,

Meets with some minor grammatical or stylistic errors.
Expectations (16- Citations and sources are mostly properly used
20 pts) and documented.
Assignment lacks clarity and structure, with

Does Not Meet many grammatical or stylistic errors. Citations
Expectations (0- and sources are improperly used or
15 pts) documented, or missing entirely.
This rubric is designed to evaluate students' ability to understand, apply,

and communicate concepts from the course in a meaningful way, as well as
their ability to critically analyse AI-generated examples. By performing well
in this assignment, students will demonstrate proficiency in many of the
key learning objectives of the course.

文件4 Interactive ChatGPT Example

Uploaded by

Copyright:

Available Formats

文件4 Interactive ChatGPT Example

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

文件4 Interactive ChatGPT Example

Uploaded by

Copyright:

Available Formats

Interactive ChatGPT Example

Generating and Evaluating Examples of Concept Transfer

To analyse AI-generated examples, assess their accuracy in demonstrating a

Transfer of knowledge on the chosen concept to various contexts, enabling

o Which aspects of the chosen concept did the AI capture correctly in

1. Task description or descriptive title:

Identify specific criteria that reflect the characteristics or components of the

3. Levels of achievement or mastery scale:

Provide descriptions of the performance quality for each criterion at each

1. Gain a thorough understanding of the fundamental concepts and theories in

Rubric for Assignment 1: Generating and Evaluating Examples of Concept

Component 1: Understanding of Concept (25%)

Chosen concept is clearly defined and explained.

Chosen concept is adequately defined and

Chosen concept is poorly defined and explained.

Component 2: Evaluation of AI-Generated Examples (25%)

Exceeds Detailed, insightful analysis provided for each AI-

concept. Proposed improvements are highly

Adequate analysis provided for each AI-generated

Limited or no analysis provided for AI-generated

Component 3: Application of Concept to Other Contexts (25%)

Exceptional demonstration of knowledge

Good demonstration of knowledge transfer, with

Limited or no demonstration of knowledge

Component 4: Clarity and Quality of Writing (25%)

Assignment is exceptionally clear, well-

Assignment is mostly clear and well-structured,

Assignment lacks clarity and structure, with

This rubric is designed to evaluate students' ability to understand, apply,

You might also like