🔢 mathematics

Critical Thinking Activation During Limits Problem Solving Mediated by Symbolab: An Instrumental Genesis Perspective

This qualitative study examines how first-year business students activate Facione's six critical thinking skills while solving calculus limits problems using Symbolab, revealing a "CT-neutral zone" of mechanical input during early instrumental genesis and identifying distinct evaluation subtypes and cross-stage patterns that bridge critical thinking theory with instrumental genesis processes.

Original authors: Enrique Ceballos-Alexis

Published 2026-06-29

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Enrique Ceballos-Alexis

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: The "GPS" Problem

Imagine you are trying to drive from your house to a new restaurant. You know the general direction, but you pull out your GPS (let's call it Symbolab) to get the exact turn-by-turn directions.

This study asks a simple question: When students use a "math GPS" to solve calculus problems, what is actually happening inside their brains?

The researcher, Enrique Ceballos-Alexis, didn't just want to know if the GPS gave the right answer. He wanted to see if the students were still thinking critically or if they just blindly followed the machine. He looked at 29 business students in Mexico solving a specific type of math problem (called "limits") using Symbolab.

The Three Stops on the Journey

The study breaks the problem-solving process into three distinct stages, like a road trip:

Stop 1: The Map Check (Before the GPS)
- What happens: The student reads the word problem and tries to figure out what math equation they need to write down before they even touch the computer.
- The Finding: Most students could understand the story (e.g., "This is about cost and distance"), but many struggled to translate that story into the correct math symbols. It's like knowing you need to go "North" but forgetting to write down the specific street name.
- The Risk: If they get the starting point wrong here, the GPS will give them a wrong answer, and they might not even realize it.
Stop 2: The "Ghost Mode" (While using the GPS)
- What happens: The student types their equation into Symbolab and waits for the step-by-step solution.
- The Big Discovery (The CT-Neutral Zone): This is the paper's most important finding. The researcher calls this the "CT-Neutral Zone."
- The Analogy: Imagine you are driving while your GPS speaks. For a few minutes, you stop thinking about where you are going or why you are taking that route. You just stare at the screen and turn the wheel exactly when the machine tells you to. Your brain goes into "autopilot."
- Crucial Detail: The study found that even very smart students (who could explain the math perfectly before using the tool) would enter this "Ghost Mode." They weren't stupid; they were just focused on the technical task of typing into the machine. Their critical thinking skills were temporarily "suspended," not broken.
Stop 3: The Destination Review (After the GPS)
- What happens: The student gets the answer and has to explain it or check if it makes sense.
- The Finding: This is where things get interesting. Some students just copied the answer. Others checked if the math steps made sense. A very few students went a step further and asked, "Does this answer make sense in the real world?" (e.g., "Wait, a negative cost for a clinic? That doesn't make sense!").

The "Three Types of Checkers"

When students did wake up from "Ghost Mode" to check the GPS's work, the study found they did it in three different ways:

The Mechanic (Procedural): "The steps look right. The numbers add up." (They checked the math, but not the meaning).
The Philosopher (Conceptual): "This result matches what I know about how numbers behave near this point." (They checked the logic).
The Realist (Contextual): "This answer says the clinic loses money, but the problem says it's a successful business. Something is wrong." (They checked the real-world story).
- Note: The "Realist" type was very rare. Most students didn't connect the math back to the story.

The "Copy-Paste" Phenomenon

One of the strangest and most telling findings was that three different students wrote the exact same sentence word-for-word when asked to evaluate the tool: "Yes, because Symbolab gives us the correct results and simplifies them."

This proves that the "CT-Neutral Zone" isn't just one student being lazy. It's a structural part of how humans interact with these tools. When we are busy typing, our brains often stop critiquing and start transcribing.

The "Broken Compass" Effect

The study found a dangerous pattern: If a student made a small mistake in Stop 1 (writing the wrong equation), they would type it into Symbolab. Symbolab would solve that wrong equation perfectly. Because the student was in the "CT-Neutral Zone" (Stop 2), they didn't question the result. By Stop 3, they confidently wrote down a wrong answer, thinking the machine was right.

What This Means (According to the Paper)

The paper does not say that tools like Symbolab are bad. Instead, it says:

Thinking is a rhythm: When we use tools, our brains naturally switch between "thinking mode" and "typing mode."
The "Neutral Zone" is normal: It is expected that students will pause their critical thinking while they are mechanically entering data. It's not a sign of stupidity; it's a sign of how our brains handle tools.
The danger zone: The problem happens when students stay in "Ghost Mode" for too long. If they never wake up to check if the answer makes sense, they might accept a wrong answer as truth.

Summary

Think of this study as a traffic report. It tells us that when students use a math GPS, they often zone out while the machine is working. The researcher built a special "observation car" (a worksheet) to watch this happen. He found that while the "zoning out" is normal, we need to make sure students wake up at the end to ask, "Does this actually make sense?" before they drive off with the answer.

Technical Summary: Critical Thinking Activation During Limits Problem Solving Mediated by Symbolab

Problem Statement
Calculus courses globally exhibit high failure rates, often attributed to shallow conceptual understanding and a reliance on mechanical algorithm application rather than reflective reasoning. While step-by-step solvers like Symbolab have become ubiquitous in student problem-solving environments, existing literature treats critical thinking (CT) in mathematics and instrumental genesis with Computer Algebra Systems (CAS) as largely independent traditions. There is a specific gap in understanding how Facione's (1990) six cognitive CT components (interpretation, analysis, inference, evaluation, explanation, and self-regulation) activate, transform, or suspend during the instrumental genesis of a CAS tool in undergraduate calculus. This study addresses the question: How do Facione's cognitive critical thinking skills activate during limits problem solving mediated by Symbolab, from the perspective of instrumental genesis?

Methodology
The study employs a qualitative, descriptive-interpretive design using reflexive thematic analysis (Braun & Clarke, 2006).

Participants: 29 first-year International Business students at the Universidad de Colima, Mexico, enrolled in a Differential and Integral Calculus course.
Instrument: A theoretically grounded, nested observation structure designed to make CT activation observable across three stages of instrumental genesis (Rabardel, 1995; Trouche, 2004):
1. Stage 1 (Pre-Symbolab): Students solve a contextualized word problem involving a cost function $C(d)$ without technology. This stage targets interpretation, analysis, and inference.
2. Stage 2 (During Symbolab): Students interact with Symbolab to generate a step-by-step solution. This stage focuses on the "CT-neutral zone" and the evaluation of the tool's output.
3. Stage 3 (Post-Symbolab): Students return to the problem to explain and self-regulate based on the tool's result.
Data Analysis: A hybrid deductive-inductive coding system was used. Deductive codes were derived from Facione's framework and instrumental genesis theory. Inductive codes emerged from the data to capture unanticipated patterns. A qualitative rubric categorized activation levels as Not Evidenced (NE), Partially Evidenced (PE), or Evidenced (E). Trustworthiness was ensured through triangulation, an audit trail, and a transparent codebook, adhering to Lincoln and Guba's (1985) criteria.

Key Findings
The analysis revealed differential activation of CT skills across the three stages, characterized by specific structural patterns:

The CT-Neutral Zone: A central construct identified is the "CT-neutral zone," a structurally determined interval during the mechanical entry of expressions into Symbolab where CT skills are transitionally suspended. This is not a sign of incompetence; rather, it is a predictable consequence of early-phase instrumentalization where cognitive resources are directed toward technical coordination. Empirical evidence includes:
- Eight of 29 participants produced literal transcriptions of Symbolab's steps without critical appraisal.
- Three distinct participants produced verbatim identical evaluation statements ("Yes, because Symbolab gives us the correct results..."), suggesting a structural phenomenon rather than individual idiosyncrasy.
- High mathematical competence in Stage 1 (e.g., precise factoring descriptions) coexisted with CT-neutral responses in Stage 2, confirming the suspension is linked to the instrumentation process, not lack of ability.
Evaluation Subtypes: Three qualitatively distinct evaluation patterns emerged inductively during Stage 2:
- Procedural (EVL-PROC): Verification of mathematical correctness without conceptual connection.
- Conceptual (EVL-CONC): Connecting steps to underlying mathematical meaning and independent checks.
- Contextual (EVL-CONT): Evaluating the result's reasonableness within the problem's real-world context (observed in only one participant).
Cross-Stage Error Propagation: Two inductive patterns describe how gaps in early stages affect later outcomes:
- ANL-OMITIDO: Skipping the analysis of structural conditions led to higher frequencies of inferential errors.
- INF-ERROR-NO-DETECTADO: Incorrect expressions entered into Symbolab were accepted uncritically, with errors propagating into Stage 3. Notably, four high-competence students accepted an incorrect limit value ($2$ instead of $-2$) without verification, illustrating how the CT-neutral zone can amplify downstream errors.
Stage 3 Activation: Explanation and self-regulation showed high variability. Self-regulation was the least frequently activated component in its full dimension. A singular case (L1-019) with prior extensive Symbolab experience demonstrated an "inverse pattern," where comprehension change was located before tool use, suggesting that advanced instrumentation allows the artifact to confirm rather than construct understanding.

Key Contributions
The study contributes an empirical-descriptive framework that systematically articulates Facione's CT components with instrumental genesis processes in undergraduate calculus. Specific contributions include:

Mathematization of CT: Adapting Facione's framework for the calculus domain, distinguishing between "CT-complete" (contextualized word problems) and "CT-truncated" (pre-formulated equations) tasks.
The CT-Neutral Zone: Providing theoretical characterization and empirical evidence for a structural interval of transitional CT suspension during tool entry, challenging the view that such pauses are merely cognitive deficiencies.
Inductive Patterns: Identifying three evaluation subtypes (procedural, conceptual, contextual) and two cross-stage error propagation patterns that extend the deductive framework.
Methodological Innovation: Utilizing a nested observation structure to make the cognitive transition from autonomous reasoning to tool-mediated reasoning and back observable, distinguishing between the mere mention of CT vocabulary and the operationalization of sub-skills.

Significance and Claims
The paper claims to be the first study to systematically describe how all six of Facione's cognitive CT components activate and transform during the instrumental genesis of a CAS tool in undergraduate calculus. It posits that the CT-neutral zone is a structural characteristic of instrumental genesis, consistent with Rabardel's (1995) framework, rather than a failure of student engagement.

The study maintains a modest scope, acknowledging limitations regarding the single institutional context, the specific topic (limits), and the reliance on written productions rather than real-time observation (e.g., think-aloud protocols). It does not claim to prove that Symbolab improves or damages CT, but rather describes the activation patterns within the mediated process. The findings suggest that task design must explicitly scaffold the analysis stage and require articulation of tool-generated steps to interrupt the mechanical-transcription state and foster critical engagement. Future work is noted as ongoing for derivatives and integrals instruments to evaluate cross-topic transformations.