Preserving Clinical Research- Daily Drip #2

 I figured we would divide today’s blog into two sections. The first part is going to talk about matrix training, with the second part talking about clinical research. Both of these are some of my favorite topics, and I would hate to miss out on either one of them. 

Matrix Training

Matrix training is nothing new. It has been around for over a decade, and is still getting updated articles about it today. However, I have never once seen it mentioned outside of the research paper. Sure, there have been modifications, updates, and advancement, but why isn’t this generalization tool a standard part of plans? I think it comes down to a couple different reasons. 

First, there is a misunderstanding regarding matrix training that I personally had and I know others have had too. It is the same misunderstanding present in multiple exemplar training. Both matrix training and multiple exemplar training is a target selection and organization tool, NOT a teaching strategy. You cannot “use” matrix training the same way you would use least to most prompting combined with error training. Matrix training is a way of selecting and organizing a massive list of targets so that the targets you teach promote more generalization relative to just presenting one after the other, error correction is a way to increase mastery of teaching targets. 

Because it does not serve as a teaching tool, it is not really absolutely necessary prior to starting an intervention. One way or another, you are going to have to tell a technician what to do following an error and a correct response. You do not, though, have to set up all your targets on a grid and work the diagonal. This extra step at the beginning, the additional response effort I think is the main reason why this method has not caught on. If you dive into the research and watch the trainings, you know that this extra 30 minutes of work saves you HOURS on the backend, but… practitioners are busy. Extra work in the moment can seem impossible, especially when it is possible to get started without drawing grids on a whiteboard. 

The other main reason I think it is not common place is the perceived complexity. “Matrix Training” sounds complicated, and the new words like “recombinative generalization,” “diagonals,” ect can seem overwhelming. At least in my grad school, there were no conversations about this. We talked for hours about how to administer a standardized assessment, but how to individually program and organize language goals? Nope. And most others I have talked to have the same experience. 

Behavior analysts are very good at using the VB-MAPP and other assessments like it to find the target (tact 30 verb-noun combinations) and then go on teachers pay teachers to snag a list of targets. Plug it into rethink or central reach and BOOM, good to go. (As you can tell, I have done this many times). 

While that method, assessment-> premade resources -> goal programming is quick and easy, what time is saved on the front end is lost because you are not programming for generalization. Generalization, the simple word that should rule over every analyst’s thought while writing a plan. “How can I do minimal programming (put the client through the least amount of hours of ABA) and get the most result (get the biggest quality of life jump).” I am not saying that matrix training is the perfect answer every time, but damn, if you can apply it to your caseload, the massive amounts of generalization that occurs is going to blow you away!

Clinical Research

Ok, let me quickly just come clean and admit that I am part of the problem. Did I complete a full blown research project and write a whole paper on publication bias with socially significant outcomes that would significantly contribute to the research base? YES. Did I, when my advisor reached out, say that I had no interest in following through with the publication process because I was too busy making content for this business and working a full time job? Also YES. Do I regret it? Eh.. a little, but not really. 

In my opinion, it is impossible to talk about the clinicians publishing research without first talking about the significant barriers associated with publication. And these are barriers that full time-researchers on grants struggle with, much less someone who has 12 clients. I won’t be able to tackle them all, but here are a couple that I personally experience as I struggle with this myself. 

• Time- Time freaking sucks. It is potentially the hardest barrier to overcome, especially for us striving to maintain some type of work life balance. As a full-time clinician, you can expect to work at bare minimum 40 hours a week on supervision, paperwork, meetings, ect. And the vast majority of us bring work home. There isn’t a lot of time to sit at the computer and type away at a research paper that potentially will be denied. Worst part is, you can’t just sub out your duties either. I cannot skip a supervision session to write a research paper. One bills out to insurance and keeps the lights on, the other… well, there isn’t really a monetary reward for it. 
• Logistics- This is another massive barrier that I have experienced. Let's say you are wanting to publish a paper involving one of your client’s skill acquisition intervention. This acquisition runs first thing in the morning, 3 days a week. However, you only have 2 hours of supervision available to spend with this client, and they are working on another 15 goals that need your attention as well. You can surely get some IOA and some treatment integrity, but it's going to be inconsistent as cancellations and plan changes inevitably happen. What about if tech calls out sick and someone covers who doesn’t know the intervention as well? Should you keep that data and eat the treatment integrity error, or should you tell them not to run the intervention that day? There are a lot of really difficult logistics barriers to overcome, especially if you are the primary writer and supervisor.

Don’t worry, I am not going to end there, complaining about all the barriers. No, I want to end on some actionable steps we can take to actually publish anyways. I have realized lately that even with all the barriers in place, we have GOT to figure out a way to publish real clinical experiences and innovations. That will be the only way to push our field towards where our client’s are begging it to go.

They say that it takes a village to raise a baby, and I think similarly regarding the research process. It takes a team to write a paper. So, the first thing to do is form your team. Sit down with your supervisor, your technician, other analysts, and talk about this project. Share with them the project and mission, and ask for support. Be upfront with your supervisor about your concerns regarding time. Maybe they can’t let you bill when you are writing the paper, but maybe they can give you from 8-9 every morning with no clients, and a locked door? Talk to your fellow supervisors about your research idea. Ask for their input and critiques, and ask them to keep an eye on any participants that may be appropriate for your project. Ask them to critique your protocols and keep an eye out for fidelity if you are not there. And finally, talk, for the love of GOD talk and involve the techs with this project. Make it a group effort for the supervisees and future analysts. Try to get techs who are passionate about ABA research to be the ones on the case. Talk to them about the importance of fidelity and help them understand the why behind the project. Odds are those techs are going to end up being second author!

All this to say, I truly am the farthest thing from an expert in this topic. I have never published a paper as a clinician, but I really want to. So please, if you have suggestions to help me in my eventual to be published journey, please dorp some comments or emails and let me know!