ChatGPT in Learning Sequences for Students

Table of Contents

I. Summary and Introductory Notes

In the article below, I’ll show you how ChatGPT by OpenAI can help you learn mathematical sequences. I present 13 examples of how ChatGPT can be used across 7 common topics related to sequences. Of course, you can mix and match your own queries to ChatGPT, like asking for problems to solve independently from any sequence topic.

What should you have or know before reading the article?

  • Access to ChatGPT, ideally version 4, which is the version I used to write this article

Educational Level

  • High School
  • University

II. Calculating Sequence Terms

Example 1 – Providing a Direct Answer to a Problem

Calculate the seventh term of the numeric sequence given by the formula a_n = {(-1)^n}n^2+3.

Tip

ChatGPT excels at interpreting text.

Write to it in your own language, use everyday vocabulary, and don’t worry too much about punctuation and grammar.

ChatGPT handles problems like this effortlessly:

You
Calculate the seventh term of the sequence ((-1)^n)*n^2+3 Copy
ChatGPT
The seventh term of the sequence (-1)^n \cdot n^2 + 3 is -46.

Example 2 – Asking for an Explanation

After getting an answer, you can ask ChatGPT for an explanation because it’s pretty good at that:

You
Calculate the seventh term of the sequence ((-1)^n)*n^2+3 Copy
ChatGPT
The seventh term of the sequence (-1)^n \cdot n^2 + 3 is -46.
You
How did you get that result? Explain. Copy
ChatGPT

To find the seventh term of the sequence given by ((-1)^n) \cdot n^2 + 3, I substituted n = 7 because we’re looking for the seventh term. This formula has two parts:

1. **Part one** is (((-1)^n) \cdot n^2), which means we raise (-1) to the power of n (in this case 7) and multiply it by the square of n, which is 7^2. (-1) raised to an odd power always results in (-1), and the square of 7 is 49. So, (-1) \cdot 49 gives -49.

2. **Part two** simply adds 3 to the result from part one, meaning we add 3 to -49.

To summarize, the calculations go like this: (-1)^7 \cdot 7^2 + 3 = -1 \cdot 49 + 3 = -49 + 3 = -46.

So, the seventh term of the sequence is -46.

Example 3 – Asking for a Solution with Explanation

How many negative terms does the sequence a_n = n^2 – 36 have?

You
How many negative terms does the sequence a_n = n^2-36 have? Explain how you arrived at the answer. Copy
ChatGPT

The sequence a_n = n^2 – 36 is a quadratic sequence where each term equals the square of the natural number n, minus 36. To determine how many negative terms this sequence has, we need to find the values of n for which the term a_n is less than 0, meaning n^2 – 36 < 0.

The quadratic equation n^2 – 36 = 0 has two solutions that correspond to the points where the sequence changes sign from negative to positive or vice versa. Let’s solve this equation:

n^2 – 36 = 0 n^2 = 36 n = \pm6

This means that for n less than 6 and greater than -6, the values of a_n will be negative. However, in the context of this sequence, we work with natural numbers n, so we’re only interested in values of n from 1 to 5 because for n = 0 and n \geq 6, the values of the sequence are non-negative. Therefore, there are 5 negative terms in this sequence, for n ranging from 1 to 5.

III. Monotonicity of the Sequence

Example 4 – Asking for a Solution with Explanation

As always, you can simply ask ChatGPT to solve the problem at any time.

Given the sequence a_n = n^2 – 4n for n \geq 4, investigate the monotonicity of this sequence.

You
Given the sequence a_n = n^2 – 4n for n >= 4, investigate the monotonicity of this sequence. Copy
ChatGPT

To investigate the monotonicity of the sequence a_n = n^2 – 4n given for n \geq 4, we’ll examine the difference between consecutive terms. If the difference is always positive, the sequence is increasing; if always negative, the sequence is decreasing; and if zero, the sequence is constant.

The difference between consecutive terms is: a_{n+1} – a_n = [(n+1)^2 – 4(n+1)] – (n^2 – 4n).

Let’s simplify this difference to determine its sign.

a_{n+1} – a_n = (n^2 + 2n + 1 – 4n – 4) – (n^2 – 4n).

a_{n+1} – a_n = n^2 + 2n + 1 – 4n – 4 – n^2 + 4n.

a_{n+1} – a_n = 2n + 1 – 4.

a_{n+1} – a_n = 2n – 3.

Now, we see that the difference between consecutive terms of the sequence is 2n – 3. To investigate monotonicity, we must determine if this difference is always positive, always negative, or can have different signs for different values of n. We’ll consider n \geq 4.

For n \geq 4, the expression 2n – 3 is always positive (since for n = 4, 2n – 3 = 8 – 3 = 5, which is positive, and increasing n only increases this value).

Thus, the sequence a_n = n^2 – 4n is increasing for n \geq 4.

Example 5 – Creating Exercises for Independent Practice

You
Please come up with 5 examples of sequences whose monotonicity I can investigate on my own Copy
ChatGPT

Here are five examples of sequences for you to investigate their monotonicity:

1. **Arithmetic sequence:** a_n = 3n + 2. Investigate the monotonicity of this sequence.

2. **Geometric sequence:** a_n = 2 \cdot 3^n. Determine if this sequence is increasing, decreasing, or constant.

3. **Quadratic function sequence:** a_n = n^2 – 2n. Examine how this sequence changes as n increases.

4. **Absolute value sequence:** a_n = |n – 5|. Determine the monotonicity of this sequence depending on the value of n.

5. **Exponential and linear function sequence:** a_n = 2^n – n. Investigate whether this sequence is increasing, decreasing, or constant based on the changes in n.

When examining the monotonicity of each sequence, the key is to understand how the difference between consecutive terms (a_{n+1} – a_n) changes or how the first derivative of a continuous function changes if the sequence is treated as a continuous function. Tip: It might be helpful to find the formula for the difference a_{n+1} – a_n and examine its sign.

IV. Recursive Sequences

Example 6 – General Theoretical Questions

You can ask Chat GPT to explain a mathematical concept:

You
What are recursive sequences? Explain it to me. Copy
ChatGPT

Recursive sequences are numerical sequences in which each term is defined using one or more previous terms of the sequence according to a certain rule, called a recursive equation. In other words, to calculate a term of the sequence, you must refer to earlier terms of the same sequence.

Recursive sequences require specifying one or several initial terms of the sequence (initial conditions) and a rule (recursive equation), which determines how to calculate each subsequent term based on the previous ones. This allows generating the sequence step by step, starting from known initial values.

Examples of recursive sequences:
1. **Fibonacci Sequence**: This is one of the most famous examples of a recursive sequence, where each term of the sequence is the sum of the two preceding ones. The recursive equation for this sequence is F_n = F_{n-1} + F_{n-2} with initial conditions F_1 = 1 and F_2 = 1.
2. **Arithmetic and Geometric Sequences**: These can also be represented recursively, where the next term of the sequence is respectively the sum or product of the previous term and a constant.
3. **Sequences Defined by Other Operations on Previous Terms**: For example, a sequence where each term is equal to the arithmetic mean, median, or another determinant of previous terms.

Recursive sequences are very useful in computer science and mathematics for modeling phenomena where each stage or state depends on the previous one, as in the case of recursive algorithms, population evolution sequences in biological models, or in finance to model various investment scenarios.

Example 7 – Practical Questions

You can also ask about things needed for practical calculations:

You
How are recursive sequences practically calculated? Is there a method for that? Copy
ChatGPT

Calculating recursive sequences in practice involves applying defined recursive equations step by step, starting from one or more initial terms. There are different approaches to calculating recursive sequences, depending on the characteristics of the sequence and available tools:

1. **Iterative Method**:

This is the most direct method, involving sequentially calculating the next terms of the sequence using the recursive equation. This usually requires a loop in the program code, which iteratively calculates each new term using the values of previous terms.

2. **Using Spreadsheets**:

Spreadsheets (e.g., Excel) can also be used, where each successive cell contains a formula referring to previous cells to calculate the next terms of the sequence.

3. **Analytical Methods**:

For some types of recursive sequences, there are methods that allow finding an explicit formula for the n-th term of the sequence without needing to calculate all previous terms. An example is the application of Binet’s formula for the Fibonacci sequence.

4. **Recursive Programming**:

In programming, recursion is a technique where a function calls itself. It can be used to calculate terms of a recursive sequence, especially when the recursive equation naturally describes such a self-calling structure. However, care must be taken with recursion depth and potential stack memory overflow.

5. **Using Mathematical Software**:

Tools such as MATLAB, Mathematica, or Python (with libraries like NumPy, SciPy) allow for effective calculation of recursive sequences, offering both iterative and analytical methods, as well as advanced features for working with recursion.

When choosing a method, consider the complexity of the sequence, available resources (e.g., software), and the purpose of the calculations. Iterative methods and recursive programming are universal but may be less efficient for large n or when an explicitly expressed form of the n-th term is available. Analytical methods and the use of specialized software are preferred for complex problems and large sequences.

Example 8 – Computer Program Code

You
Given a sequence a_n defined by the recursive formula a_1 = 1, a_(n+1) = 10-2a_n. Write a Python code to find the 250th term of this sequence. Copy
ChatGPT

It seems there was a technical issue while trying to perform the calculations. However, I can show you how to write and run similar code on your own computer. Here’s an example of Python code that calculates the 250th term of a recursive sequence a_1 = 1, a_{n+1} = 10 – 2a_n:

def compute_nth_term(n): a_n = 1
# Initial condition
for _ in range(1, n):
a_n = 10 – 2*a_n
# Recursive equation
return a_n
# Calculate the 250th term of the sequence
nth_term = compute_nth_term(250)
print(nth_term)

This code defines a function `compute_nth_term`, which calculates the n-th term of the sequence using the given recursive formula. It then calls this function for n = 250 and displays the result. If you want to find out what the 250th term of this sequence is, you can run this code in your programming environment.

V. Arithmetic Sequences.

Example 9 – Formulas

You
Give me the formulas needed for calculations related to arithmetic sequences. Copy
ChatGPT

Arithmetic sequences are one of the simplest types of numerical sequences, in which the difference between any two consecutive terms is constant. Here are the basic formulas used in calculations related to arithmetic sequences:

1. **Formula for the n-th term of an arithmetic sequence**:

a_n = a_1 + (n – 1)d
where:
a_n is the n-th term of the sequence,
a_1 is the first term of the sequence,
d is the common difference (constant value added to each term to get the next one),
n is the term number.

2. **Formula for the sum of the first n terms of an arithmetic sequence**:

S_n = \frac{n}{2} \left(2a_1 + (n – 1)d\right)
or equivalently:
S_n = \frac{n}{2} \left(a_1 + a_n\right)
where:
S_n is the sum of the first n terms of the sequence,
– other symbols are as defined above.

These two basic formulas allow solving most problems related to arithmetic sequences, such as finding specific terms of the sequence, calculating the sum of the sequence, or determining properties of the sequence based on its terms.

VI. Geometric Sequences

Example 10 – Real-World Application Question

You
What are the real-world applications of geometric sequences? In what fields are they used? Copy
ChatGPT

Geometric sequences, characterized by the fact that each term of the sequence (except the first) is the product of the previous term and a constant called the ratio, find wide applications across various fields of science, engineering, finance, and everyday life. Here are some practical examples of their applications:

1. **Finance and Economics**:

– Calculating compound interest, where the value of a deposit or loan grows by a fixed percentage each period.
– Modeling economic growth or inflation over time.
– Investment analysis, including future value calculations of payment streams or annuities.

2. **Biology and Medicine**:

– Modeling population growth in environments where the number of individuals in the next generation is proportional to the current number.
– The spread of epidemics, where the number of infected individuals may grow exponentially in the early stages.

3. **Physics and Chemistry**:

– Calculations related to radioactive decay, where the amount of substance decreases by a fixed percentage each half-life period.
– Acoustic and optical phenomena, e.g., the attenuation of light or sound intensity in a medium according to a geometric law.

4. **Computer Science**:

– Algorithms with logarithmic or exponential complexity, where execution time is a geometric function of input data size.
– Data storage in tree-like structures, such as binary trees, where the number of nodes at each level grows geometrically.

5. **Technology and Engineering**:

– Signal and systems analysis, where patterns of signal decay or amplification can be modeled using geometric sequences.
– Designing electrical and acoustic networks, where values of components (e.g., resistance, capacitance) may be arranged in geometric sequences for optimal characteristics.

Geometric sequences are thus extremely useful for modeling processes that develop or decrease proportionally, making them an important tool in many scientific and technical fields.

Example 11 – Word Problem

ChatGPT handles word problems excellently:

Marek kicked a ball upwards with all his might. The ball rose and fell, bouncing off the court. Calculate how high the ball rose the fourth time, if the first time it rose to 10 meters, and then to half the height it had reached the previous time.

You

Solve the word problem:

Marek kicked a ball upwards with all his might. The ball rose and fell, bouncing off the court. Calculate how high the ball rose the fourth time, if the first time it rose to 10 meters, and then to half the height it had reached the previous time. Copy

ChatGPT

To calculate how high the ball rose the fourth time, we can use the fact that the ball rises to half the height of the previous time on each subsequent rise. This means we are dealing with a geometric sequence where the first term a_1 = 10 meters, and the common ratio q = \frac{1}{2}.

The formula for the n-th term of a geometric sequence is:
a_n = a_1 \cdot q^{n-1}

We want to find a_4, so we substitute n = 4, a_1 = 10, and q = \frac{1}{2}:
a_4 = 10 \cdot \left(\frac{1}{2}\right)^{4-1}

Let’s calculate that.

It appears there was a technical issue while trying to perform the calculation. Nonetheless, we can easily compute this manually using the given formula:

a_4 = 10 \cdot \left(\frac{1}{2}\right)^{3} = 10 \cdot \frac{1}{8} = 1.25

This means that the ball rose to a height of 1.25 meters the fourth time.

VII. Limits of Sequences

Example 12 – Request to Confirm Own Calculations

You can present your calculations to ChatGPT and ask for verification:

Calculate the limit of the sequence {a}_{n} = \sqrt[n]{2^n+3^n+4^n}.

You

Solve the problem:

While calculating the limit of the sequence (2^n+3^n+4^n)^(1/n) using the squeeze theorem, I chose the lower bounding sequence: (4^n)^(1/n) and the upper bounding sequence: (4^n+4^n+4^n)^(1/n). Did I choose these sequences correctly? Copy

ChatGPT

Yes, your approach to applying the squeeze theorem (also known as the sandwich theorem or the two policemen theorem) is correct in the context of this task. This theorem states that if the sequence b_n is bounded below by sequence a_n and above by sequence c_n, and the limits of sequences a_n and c_n as n approaches infinity are equal, then the sequence b_n has the same limit.

In your case, the sequence is given by:
\left(2^n + 3^n + 4^n\right)^{\frac{1}{n}}

You chose the lower bounding sequence:
\left(4^n\right)^{\frac{1}{n}} = 4

and the upper bounding sequence:
\left(4^n + 4^n + 4^n\right)^{\frac{1}{n}} = \left(3 \cdot 4^n\right)^{\frac{1}{n}} = 3^{\frac{1}{n}} \cdot 4

Calculating the limit of both bounding sequences as n \to \infty, we get:

For the lower sequence (bounding below):
\lim_{n \to \infty} 4 = 4

For the upper sequence (bounding above):
\lim_{n \to \infty} 3^{\frac{1}{n}} \cdot 4 = 4

Since the limits of both bounding sequences are equal and equal to 4, by the squeeze theorem, the limit of the sequence:
\lim_{n \to \infty} \left(2^n + 3^n + 4^n\right)^{\frac{1}{n}}
also equals 4.

Your reasoning is therefore correct. You applied the squeeze theorem properly by choosing the appropriate bounding sequences, which allowed you to find the limit of the given sequence.

Example 13 – ChatGPT Mistake

Unfortunately, ChatGPT is not a perfect tool. It can make mistakes, for example in more complicated calculations.

Calculate the limit of the sequence a_n = \frac{5^n-2^n+10^n}{11^n+5^n}

You
When calculating the limit of the sequence step by step, I made the following transition: lim_(n->oo) (5^n-2^n+10^n)/(11^n+5^n) = lim_(n->oo) (10^n((5/10)^n-(2/10)^n+1)/(11^n(1+(5/11)^n)). Did I make this transition correctly? Copy
ChatGPT

Your transition when calculating the limit of the sequence looks like an attempt to simplify the expression by factoring out the largest power of n in the numerator and denominator, which is a common method when calculating limits of sequences with exponential expressions. However, let’s check the accuracy of your transition:

The initial expression is:
\lim_{n \to \infty} \frac{5^n – 2^n + 10^n}{11^n + 5^n}

Your transition to:
\lim_{n \to \infty} \frac{10^n \left(\left(\frac{5}{10}\right)^n – \left(\frac{2}{10}\right)^n + 1\right)}{11^n \left(1 + \left(\frac{5}{11}\right)^n\right)}

It appears that you made a mistake by factoring out the common factor 10^n in the numerator without properly accounting for the terms 5^n and 2^n, and by trying to simplify the denominator by factoring out 11^n without maintaining consistency with the numerator transformation.

The correct way to factor out the largest power of n should focus on the largest power in both the numerator and the denominator to facilitate limit analysis. In this case, the largest power of n in the entire expression is 10^n in the numerator and 11^n in the denominator. The correct transformation should clearly separate these terms and simplify the rest appropriately, which might look something like this:

\lim_{n \to \infty} \frac{10^n + 5^n – 2^n}{11^n + 5^n} = \lim_{n \to \infty} \left(\frac{10^n}{11^n} + \frac{5^n}{11^n} – \frac{2^n}{11^n}\right)

Then analyzing each component separately. However, your transformation introduces errors in the form of improper use of the common factor and incorrect simplification of the expression. The correct direction is to consider the relative size of the terms of the expression to the largest power of n in the expression, comparing each term to 10^n or 11^n and analyzing their behavior as n approaches infinity.

Tip

Yes, ChatGPT can simply make mistakes. This is all the more dangerous because it does not admit them. Instead of stating that it doesn’t know something, it prefers to fantasize. Therefore, you cannot rely on it 100%.

In addition to ChatGPT, use mathematical calculators and applications such as WolframAlpha, or MathKiwi.

My transition was correct, although ChatGPT stated that it was not correct.

What’s even more disappointing, it tried to support its reasoning with vague and protracted considerations, or as it is termed – “hallucinating.”

Bibliography

In writing this article, I used:

Picture of Krystian Karczyński

Krystian Karczyński

Founder and head of the mathematical services MathKiwi and eTrapez.

Graduate of Mathematics at the Poznań University of Technology. Math tutor with many years of experience. Creator of the first eTrapez Courses, which have become extremely popular among students across Poland.

Lives in Szczecin, Poland. Enjoys walks in the forest, beach outings, and kayaking.

Leave a Reply

Your email address will not be published. Required fields are marked *