Co-Teaching Models That Actually Work for High School Students With Severe Learning Disabilities
Most schools default to "one teach, one assist" because it's the easiest to schedule. The supporting teacher floats around helping kids who are stuck while the lead teacher covers content. It sounds cooperative but in practice it usually means the student with severe learning disabilities gets treated like an afterthought. They don't get differentiated instruction. They get whispered remediation at their desk while the rest of the class moves forward. I've seen it happen every semester for twelve years. The model I recommend, and what I've built my practice around, is a hybrid of station teaching and alternative teaching. It's not elegant. It requires genuine coordination between two educators who actually like each other. But it produces results where the other models consistently fail. Here's how it works in practice. You divide the class into two groups. One group stays with the general education teacher working through the core lesson at an accessible pace. The other group rotates to the special education teacher for a targeted, scaffolded session on the same material. Then you switch. Every twenty to twenty-five minutes. You use a visible timer. Students know exactly what's coming next. Predictability reduces anxiety, and anxiety is a major blocker for students with severe learning disabilities.
The critical detail that everyone misses: the content must be identical across both stations. Not similar. Not aligned. Identical. If the general ed teacher is covering main ideas from a history chapter, the special ed teacher is covering the exact same main ideas, just broken down with visual supports, simplified language, and chunked instructions. When they diverge, the students who rotate notice immediately. They feel like they're getting the easy version. That damages engagement fast. I ran into a specific problem last year that made me rethink how I handle the transition between stations. My student with severe dyslexia and an IQ in the borderline range would completely shut down during the switch. Not because he didn't understand the work. He understood it fine. It was the context shift. Moving from one room to another, sitting at a different table, switching from one teacher to another — it overloaded his working memory. He'd lose the thread entirely and then refuse to re-engage because re-engaging felt like starting over from zero. My workaround was brutal but effective. I stopped moving the students at all. Instead, I moved the teachers. The general education classroom has enough space if you clear out the desks. I physically moved my teaching station — whiteboard, materials, projected slides — to the special education corner of the room for the alternative teaching block. The students stayed seated. They only changed which adult was addressing them. The visual continuity of the room and the routine of the space kept him anchored. It cut transition-related off-task behavior from roughly eight minutes per rotation to about ninety seconds.
Another nuance that isn't talked about enough: the ratio of instruction time matters more than the quality of any single lesson. For severe learning disabilities in high school, I've found that 15 to 20 minutes of focused, alternative instruction per rotation is the sweet spot. Anything longer and retention drops off a cliff. Their working memory capacity simply doesn't sustain that kind of cognitive load. But two short rotations per class period — that's 30 to 40 minutes of targeted support instead of the traditional model's five minutes of desk-side whispering. The math is undeniable. The alternative teaching component — pulling a small group out for explicit remediation — works well for specific skill deficits. I use it for students who need intensive phonics work, for those processing numbers, or for kids who need sentence-level writing support before they can tackle paragraph structure. The key is that this isn't "easy work." It's harder work, just at a more accessible entry point. I've seen schools use alternative teaching as a dumping ground for busy work while the gen-ed teacher covers real content. That's a failure of implementation, not a failure of the model. Team teaching — both educators up front delivering the lesson together — is attractive on paper but practically nearly impossible with severe learning disabilities unless both teachers are highly experienced and have practiced the coordination extensively. The cognitive load of managing dual delivery while simultaneously differentiating for students who need constant redirection is too high. I tried it for one semester. It was chaotic. The students with severe disabilities, who need the most structure, got the least. I switched back to the station-alternative hybrid and saw improvement within three weeks.
Get the Full Details
One Teach, One Observe sounds useful for professional development. In reality, the observing teacher often ends up taking notes for administrative compliance rather than gathering actionable data. And the student with severe learning disabilities doesn't benefit from being observed — they benefit from being instructed. I repurposed the observation model by having the observing teacher actively take frequency data on target behaviors instead of generic notes. That produced usable IEP progress data and gave the lead teacher real-time information about which students were disengaging and when. Parallel teaching — splitting the class in half and each teacher delivering the same lesson to their own group — is essentially station teaching without the rotation. It's faster to set up. It works if your class size is twenty-four or fewer and both teachers are comfortable with the content. But it gives each student with severe learning disabilities only half the exposure time compared to the rotating model. For kids who need repetition and reinforcement, that half-dose is often the difference between mastery and confusion. The biggest barrier to implementing any of this isn't pedagogy. It's scheduling. Co-planning time. Common planning periods. If your general education teacher and special education teacher never share a planning period, the station model becomes a logistical nightmare. I've seen teams make it work by rotating the schedule every six weeks so each teacher gets a common planning block at some point during the semester. It's not ideal. It's better than nothing.
Another thing nobody prepares you for: the paraprofessional question. In many high school settings, you'll have apara assigned to support students with severe disabilities. If they're not trained in the co-teaching model, they become a crutch. The student learns to wait for the apara to re-explain everything instead of engaging with the teacher. I make it a non-negotiable rule that paras do not re-teach. They prompt. They redirect. They scaffold the process, not the content. The teachers deliver the instruction. This takes discipline from the paras and the administrators who assign them, but it prevents learned helplessness, which is a far more common problem than anyone wants to admit. Here's what I'd tell someone walking into this for the first time. Start with the station-alternative hybrid. Keep it simple. Two groups. Twenty-minute rotations. Identical content. A visible timer. Don't try to layer in team teaching or parallel teaching until your team has executed the basic model successfully for at least six weeks. Document what's working and what isn't. Adjust the rotation length based on your students' actual attention spans, not a textbook recommendation. Some kids with severe disabilities need fifteen-minute blocks. Some can handle thirty. Let the data tell you. The model isn't perfect. It demands more preparation from both teachers. It requires the school to provide adequate space and a workable schedule. It doesn't help if one teacher is actively resistant to sharing instructional responsibility. There are days when the rotation falls apart because a student is having a rough day and won't comply with the switch. On those days, you pivot to whatever keeps them learning, even if it means abandoning the model temporarily.
But compared to the alternatives — especially the default of one teacher dumping content at twenty-five different learning levels simultaneously — it's the best option available. The students who need the most structure and the most repetition get it. The ones who can handle the general pace stay engaged with grade-level content. And both teachers are actually teaching instead of one teaching while the other watches or assists superficially.
