In my years as a math teacher and tutor, I noticed a pattern I couldn’t quite explain.
Students who struggled with math anxiety often had parents who nervously apologized before parent-teacher conferences: “I was never good at math either.” But that didn’t fully account for what I was seeing. Some of these parents were actually quite capable with numbers—they managed household budgets, calculated tips, helped with homework through middle school. Their anxiety wasn’t about doing math. It was about something else.
It took me years to understand what I was observing: These parents weren’t anxious about math itself. They were anxious about their child’s relationship with math—the grades, the future implications, the worry that their child might be “falling behind.” And somehow, that adult anxiety was reaching their kids, even when the parents never said a word about their fears.
Now, new research published in January 2026 has finally put a name to what I witnessed in countless classrooms: Parental Mathematics Education Anxiety (PMEA)—and it’s fundamentally different from the math anxiety we typically talk about.
But here’s what matters most: This isn’t just about parents. It’s about how children learn to feel about anything uncertain—including AI, which is the newest “math” on the block.
A New Framework for Understanding How Adult Anxiety Affects Learning
A comprehensive study from Henan University involving 260 student-parent pairs made a critical distinction that changes how we understand math anxiety transmission.
The researchers identified three types of anxiety in adults:
- General State Anxiety – baseline nervousness/worry about life in general
- Math Anxiety (PMA) – personal discomfort with doing math themselves (“I’m not a math person”)
- Parental Mathematics Education Anxiety (PMEA) – situational anxiety specifically about their child’s math education
Here’s what the study found: After controlling for general anxiety and parents’ own math anxiety, PMEA was the only significant predictor of children’s math achievement. The adult’s personal math ability or general anxiety levels didn’t matter nearly as much as their anxiety about how their child was doing in math.
What PMEA Actually Looks Like
PMEA isn’t “I can’t do algebra.” It’s:
- The knot in your stomach when your child brings home a C in math (but not in English)
- The frustration you feel when homework time stretches to two hours because they “just don’t get it”
- The worry that spirals after a parent-teacher conference: What if they can’t handle calculus? What does this mean for college?
- The defensiveness when family members ask about grades or compare your child to cousins
- The headache that comes when you even think about sitting down for math homework
A parent with high PMA might say: “I hate math.”
A parent with high PMEA might say: “I’m fine with math, but watching my kid struggle with it is driving me crazy.”
The distinction matters because it changes who’s affected and what needs to be addressed.
You might be perfectly comfortable balancing a budget or calculating percentages—but still experience intense PMEA when your child’s math grades slip. Conversely, you might have significant math anxiety yourself but remain calm and supportive about your child’s learning process, keeping PMEA low.
The Study’s Key Finding
The experimental phase of the research tested whether positive encouragement interventions would help children’s math performance. The results revealed something unexpected:
For children without math learning difficulties:
- Children of low-anxiety adults who received frequent encouragement → improved performance
- Children of high-anxiety adults who received frequent encouragement → worse performance (the “backfire effect”)
For children with math learning difficulties:
- In low-anxiety homes, minimal support was highly effective
- In high-anxiety homes, intensive high-frequency support was necessary for gains
The takeaway: It’s not what you say. It’s the emotional state you’re in when you say it.
How Children Learn to Feel: The Science of Social Referencing
The transmission of PMEA isn’t mysterious—it’s rooted in a well-established developmental phenomenon that child psychologists have studied for decades: social referencing.
What Is Social Referencing?
Starting around 6-9 months of age, children begin using facial expressions, vocal tones, and body language from trusted adults to determine how they should feel and respond in ambiguous or novel situations.
It’s an evolutionary survival mechanism: I don’t know if this is dangerous. Let me check with the person who keeps me alive. Oh, they look scared? I should be scared too.
The Visual Cliff Experiment
One of the most famous demonstrations of social referencing is the “Visual Cliff” study. Researchers created an illusion of a drop-off using clear glass over a patterned floor. When babies crawled to the edge and looked down at the “cliff,” they would stop and look back at their mother.
If the mother smiled and looked encouraging, the baby would crawl across the glass (despite the visual illusion of danger).
If the mother looked fearful or worried, the baby would refuse to cross—even though the surface was completely safe.
The baby didn’t need verbal instruction. They read the emotional data and made a decision.
Social Referencing Doesn’t Stop in Infancy
While the mechanism is most obvious in babies, social referencing continues throughout childhood and even into adolescence. Children—especially when facing something new, uncertain, or evaluative—constantly scan trusted adults for emotional cues:
Is this safe? Is this something I should worry about? Am I doing okay? Should I feel proud or ashamed?
Trusted adults include:
- Parents and caregivers
- Teachers and coaches
- Grandparents
- Mentors
- Anyone the child perceives as competent and caring
This means PMEA isn’t just a parent problem—it’s an ecosystem problem.
If a child’s teacher shows visible frustration during math instruction, sighs heavily when students struggle, or makes comments like “This should be easy by now,” the child absorbs that emotional data.
If a coach dismisses math as unimportant (“You don’t need that stuff in the real world anyway”), the child learns that math isn’t worth investing in.
If a tutor shows impatience or anxiety about the child’s progress, the child interprets struggle as evidence of inadequacy rather than a normal part of learning.
The Multimodal Transmission
The research on PMEA found that adult anxiety transmits through multiple channels simultaneously:
- Facial expressions: Tightness around the eyes, furrowed brows, forced smiles that don’t reach the eyes, looks of disappointment
- Vocal tone: Impatience, tension, forced cheerfulness, sighs, sharpness even when words are kind
- Body language: Rigid posture, hovering, tapping fingers, checking the clock, crossing arms
- Verbal cues: “This is taking too long.” “You should know this by now.” “Why don’t you get this?”
Children are extraordinary emotion-readers. They can detect incongruence—when your words say “It’s okay to make mistakes” but your body language says “This is taking too long and I’m stressed.”
And here’s the critical insight: You can’t fake calm. Children don’t respond to what you’re trying to project. They respond to what you’re actually feeling.
This is how student math anxiety often begins—not with their own struggles, but with absorbing the emotional climate created by the adults around them.
AI Anxiety: What Happens When Everyone’s Standing at the Edge of a New Cliff
Right now, children are encountering a new “visual cliff” in their educational lives: artificial intelligence.
Many students are already using AI tools on their own—ChatGPT for homework help, character.ai for conversation practice, AI image generators for creative projects. But when it comes to AI in the context of school and learning, they’re looking to adults for emotional cues:
Is this cheating? Is this dangerous? Am I allowed to use this? Should I feel guilty? Excited? Worried about my future?
And what are they seeing when they look at the adults around them?
Teachers standing at their own cliff, looking anxious.
In a previous article, I explored causes of teacher AI anxiety —and it’s not primarily about job security or lack of technical skills. Teachers are experiencing something remarkably similar to PMEA: AI Education Anxiety.
They’re anxious about:
- Whether they’re preparing students adequately
- How to detect and respond to AI use in assignments
- Whether their curriculum is becoming obsolete
- The ethical ambiguity of a tool that didn’t exist when they were trained
- The pressure to “integrate AI” without clear guidance on how
And students are absorbing that anxiety.
When a teacher:
- Bans AI with harsh warnings about cheating
- Expresses frustration about “kids taking shortcuts”
- Admits they don’t understand how to use these tools
- Shows visible stress when AI use is suspected
Students learn: AI is threatening. I should feel anxious about it. Using it means I’m doing something wrong.
This creates the same transmission loop that the PMEA research identified:
- Adult experiences situational anxiety about student’s relationship with [math/AI/new technology]
- Adult attempts to manage that anxiety through control, warnings, or over-involvement
- Student perceives the anxiety and interprets it as: This is dangerous/I’m not capable/I should be worried
- Student’s performance or relationship with the subject deteriorates
- Adult’s anxiety increases
- Cycle repeats
The Parallel Is Striking
Just as parents with high PMEA can make math worse through well-intentioned but anxiety-laden support, teachers with high AI anxiety can make students more fearful and less equipped to navigate an AI-integrated world.
The solution isn’t to pretend we’re not uncertain. It’s to model how to stand at the edge of uncertainty without transmitting panic.
The Hidden Factor in the Achievement Gap
One of the study’s most significant findings has profound implications for how we understand educational equity.
The researchers found that PMEA was significantly higher in:
- Families with lower household income
- Parents with lower educational attainment
At first glance, this might seem to confirm existing narratives about the “achievement gap.” But the mechanism the researchers identified challenges common assumptions about why low-income students often have lower math performance.
It’s Not About Resources—It’s About Optionality
We’ve long attributed disparities in math achievement to resource gaps: wealthier schools have better technology, smaller class sizes, more experienced teachers, advanced courses, tutoring access.
And yes, those factors matter. But this research suggests another powerful variable we’ve underestimated: the ambient stress level in which learning occurs.
The researchers explain the income-PMEA correlation through what I’ll call reduced optionality:
When families perceive education as the only viable pathway to economic security, every academic stumble feels existential.
Example: Two Students, Different Stakes
Consider two seventh-graders both struggling with algebra:
Student A (lower-income family):
- Parents working multiple jobs with limited time for homework help
- Family stress about bills, housing stability, healthcare costs
- College will require significant scholarships
- Career options feel limited without a degree
- Academic performance = the narrow bridge to a different future
The parent’s internal monologue: If they don’t get this, they won’t qualify for honors math. Without honors math, college applications are weaker. Without college, they’ll face the same economic struggles we do. Every homework session matters. Every grade matters. I can’t let them fail.
Student B (higher-income family):
- Parents with flexible schedules or ability to hire tutors
- Financial stress is minimal
- College is assumed and financially covered
- Family has professional networks that could provide career opportunities regardless of GPA
- Multiple pathways to success are visible
- Academic performance = one of many indicators, not the sole determinant
The parent’s internal monologue: Math is hard for them right now, but they’ll figure it out. We can get a tutor if needed. Even if they don’t excel in math, they’re creative and social—plenty of paths forward.
The Anxiety Differential
Student A is absorbing significantly more ambient stress during math homework, not because their parent is “worse” at parenting, but because the structural stakes are higher.
The parent of Student A is experiencing justified anxiety—the fear isn’t irrational. In their economic reality, academic performance does have higher consequences. The safety net is thinner. The margin for error is smaller.
This reframes the achievement gap question:
It’s not: “Why can’t low-income students perform as well?”
It’s: “How can any student perform well when they’re doing cognitive work under high ambient stress?”
The research showed that anxiety occupies working memory, slows information processing, and reduces the cognitive resources available for mathematical problem-solving. When a child is simultaneously:
- Trying to understand a fraction problem
- Managing their own frustration
- Reading their parent’s stress
- Worrying about consequences of failure
…their brain simply has less capacity for the math itself.
Why This Matters for Educators
If you’re a teacher in a Title I school or working with predominantly lower-income students, this research suggests your students may be arriving to math class already carrying a heavier emotional load.
The “achievement gap” may be partially an anxiety gap—one rooted in structural economic realities, not deficits in ability or effort.
This doesn’t mean lowering standards. It means recognizing that emotional accessibility must come before cognitive challenge. Before you can teach effectively, you may need to actively reduce the ambient stress in the room.
How to Be the Calm Adult in the Room
Whether you’re a parent, teacher, tutor, coach, or any other adult working with young people, the research offers clear guidance: Your emotional state matters more than your strategies.
Here’s how to apply these findings across different contexts:
For Parents: Managing PMEA at Home
1. Identify the Source of Your Anxiety
PMEA doesn’t exist in a vacuum. Ask yourself:
- Am I anxious about math performance specifically, or defending my educational choices to others?
- Is this about grades, or about what grades mean for their future?
- Am I comparing my child to siblings, cousins, neighbors?
- Is my anxiety actually about my own economic insecurity?
Naming the source helps separate your anxiety from their learning.
2. Adjust Your Involvement Based on Your Emotional State
The research showed that high-frequency positive suggestions from high-anxiety adults can backfire. This means:
If you’re highly anxious:
- Less frequent, genuinely calm interaction is better than frequent anxiety-laden “support”
- Consider whether another adult (partner, grandparent, tutor) with lower anxiety could handle homework help
- Focus on autonomy-supportive language: “What strategy do you want to try?” vs. “Let me show you the right way”
If you’re low-anxiety:
- Consistent, frequent positive encouragement is highly effective—lean into it
- Your calm presence is the intervention
3. Practice Emotional Transparency (When Appropriate)
You don’t have to pretend you’re never worried. But you can model how to hold worry without transmitting panic:
❌ Don’t say: “This is so frustrating! Why don’t you get this?”
✅ Do say: “I notice I’m feeling some tension right now—that’s about me, not about you. Let’s take a break and come back to this.”
This teaches emotional regulation while preventing math anxiety transmission to your child.
For Teachers: Creating Low-Anxiety Math Classrooms
1. Recognize You’re Modeling Emotional Response to Struggle
Students are constantly reading your face, tone, and body language when they hit difficulty. They’re asking: Is it okay that this is hard? Is struggle normal or shameful?
What helps:
- Normalize struggle explicitly: “This problem should feel tricky—that means your brain is working.”
- Monitor your facial expressions when students make errors
- Celebrate productive mistakes: “That wrong answer just taught us something important.”
2. Create “Safe Struggle” Zones
The research found that adult anxiety reduces students’ working memory capacity. To counteract this:
- Use low-stakes practice before high-stakes assessment
- Offer multiple attempts on assignments
- Provide reassurance that grades reflect learning progress, not worth
- Build in “thinking time” without pressure to perform quickly
3. Address AI with Curiosity, Not Fear
If students are reading your AI anxiety, they’re learning to fear it rather than navigate it skillfully.
What helps:
- Admit uncertainty without panic: “I don’t have all the answers about AI yet—we’re figuring this out together.”
- Model ethical decision-making: “Here’s how I think through when to use AI and when not to.”
- Frame AI as a tool requiring judgment, not a threat or a cheat code
For Anyone Working with Kids: Universal Principles
1. You Can’t Fake Calm—So Get Actually Calm
Children are too good at reading emotional incongruence. If you’re anxious, they’ll know—even if you smile and say the right words.
This means:
- Do your own emotional work (therapy, journaling, talking to other adults)
- Take breaks when you feel frustration rising
- Don’t force yourself to “power through” high-stress interactions
2. Prioritize Emotional Safety Over Performance
The study proved that in low-anxiety environments, even minimal support is effective. The emotional climate is the foundation—not the add-on.
Ask yourself:
- Does this child feel safe making mistakes in front of me?
- Do they believe I see their struggle as normal or as disappointing?
- Am I communicating (verbally or non-verbally) that their worth depends on their performance?
3. Remember: They’re Watching You Navigate Uncertainty
Whether it’s math, AI, or the future economy, children are looking to adults to see:
- Is uncertainty terrifying or manageable?
- Is it okay to not know something?
- Can I try new things even when I’m not sure I’ll succeed?
The way you stand at the edge of the cliff—calm, curious, careful but not panicked—teaches them how to navigate their own.
The Adults in the Room
When I look back at my years teaching math, I realize I wasn’t just teaching algebra or geometry. I was teaching students how to feel about not understanding something yet.
And they were learning that not just from me, but from every adult they trusted—parents worried about grades, coaches dismissing academics, tutors rushing through explanations with barely concealed frustration.
This new research on PMEA confirms what I observed but couldn’t name: Anxiety is contagious. And it doesn’t need words to spread.
The most powerful thing we can do for young people—whether we’re teaching math, introducing AI, or helping with homework—isn’t to have all the answers. It’s to manage our own emotional state so they don’t have to.
That doesn’t mean pretending we’re not uncertain or worried. It means modeling how to hold uncertainty without panic, how to face struggle without shame, how to stand at the edge of something new and trust that we’ll figure it out together.
Children don’t need perfect adults. They need adults who can be present without adding their own unprocessed anxiety to the child’s learning experience.
They’re watching us. Not just for answers—but for how to feel about the questions.
Want to assess your own anxiety around math or AI education and develop strategies for managing it? Download our free guide: “The Calm Adult Framework: Managing Your Anxiety So Kids Don’t Have To.”