In conversation with Sarah Wood about Assistive Technology

In this episode, you will listen to and learn from Dr. Sarah Wood. She is a cognitive scientist and assistive technology expert with a passion for helping people with disabilities. She obtained a Ph.D. in Developmental Psychology from Florida State University and was a research assistant at the Florida Center for Reading Research. Currently, she is an Accessibility Scientist at Educational Testing Services, better known as ETS. Her passion is leveraging her unique set of talents to empower all learners using her expertise in research, text-to-speech, and accessibility technology more generally. Through her advocacy, research, and systems thinking, she is an emerging leader in efforts to facilitate the learning process so that people with disabilities can reach their highest potential! Especially how people with disabilities can access their education and find how technology works best for them.

Continue the Learning

Read Sarah's Meta-Analysis on the effect of Text-to-Speech on Reading Comprehension.

Learn about Microsoft Immersive Reader

Learn about Grammarly

Learn about Equation Editor from texthelp

Learn about Microsoft Copilot

Learn about ChatGPT

The conversation highlights these assistive technologies and AI solutions. There are many others to explore and learn about.

Share your thoughts and follow your host on X @OdegardTim

Transcript

Dyslexia Uncovered in Conversation with Sara Wood about Assistive Technology
===

[00:00:00] Intro Music 

Tim Odegard: Welcome Dyslexia Uncovered community. How are you guys doing today? It's been a while since I reached out and shared an episode and well, life happens. There's been a lot going on and I wanted to start off by taking a moment to thank you. You guys have been listening. You've been sharing and you've even reached out to tell me how much this podcast has meant to you.

Tim Odegard: Thank you. It means so much to hear from you and learn. How its touched you I learned about going on a walk with someone who I don't even know because she brought me and Maryanne with her as she went through the woods. I helped someone's fitness because they were really curious and they wanted to listen to all the episode.

Tim Odegard: So they [00:01:00] walked an extra mile on their walk. Well, you know, I'm here for lots. If I'm your fitness partner, that's great too. And I've heard about those driveway moments you guys are having where you're lingering a little bit longer as you wrap up your commute. So that you can listen to just a little bit more of the episode.

Tim Odegard: So thank you. It means so much to me. Today is a special conversation. It's with an emerging thought leader in the area of assistive technology, and she happens to be a friend of mine who I've known quite a while now. But maybe just as importantly, it's the first conversation with someone else like me who grew up with similar differences, if you will, like me, because she too has dyslexia.

Tim Odegard: And it's nice to sit in community with someone else who has those similar experiences, who is dealing with similar obstacles and challenges, and is trying out new [00:02:00] things like assistive technology for herself to figure out how it works for her. So today is a conversation with someone who shares her deep knowledge about assistive technology and some of the research she's done.

Tim Odegard: She talks about some of the trends she sees coming, as well as We talk about how we use technology in our daily workflows, something I really said I wanted to highlight, and I'm glad I got to do it with her. So today you will listen and learn from Dr. Sarah Wood. She's a cognitive scientist and assistive technology expert with a passion for helping people with disabilities.

Tim Odegard: Now, I know that disabilities is a loaded word. Maybe we should say neurominorities, but she uses the word disabilities. We could also use neurominorities. She obtained a PhD in developmental psychology from Florida State University and was a research assistant at the Florida Center for Reading Research.

Tim Odegard: Currently, she is an accessibility [00:03:00] scientist at Educational Testing Services, better known to many of you as ETS. Her passion is in leveraging her unique talents to empower all learners using her expertise in research, text to speech, and accessibility technology more generally. Through her advocacy, research, and systems thinking, she is an emerging thought leader in efforts to facilitate the learning process for people with disabilities to reach their highest potential, especially how people with disabilities can access their education and find how technology best works for them.

Tim Odegard: What struck me as I reflected on this conversation, or put better, spent hours upon hours upon hours editing this conversation, was. How much humility Sarah brought to it as she shared so much knowledge about a critical topic. For our community, one of the things that I want each of you to learn [00:04:00] from this podcast is that each of us can do far more than what society may think we can.

Tim Odegard: And Sarah and I talk about this. We share that each of us in our own way had people tell us what we couldn't do or what wouldn't be available for us. But none of those people ever once thought that either of us would grow up to reshape how this world works. But here we are. And we happen to do that. Now in the conversation because of our personalities, neither of us reflected on that ourselves. Sarah actually points to how the younger generations of people like us, neuro minorities are much more vocal and what their needs are and how they are reshaping the world. But Sarah is doing that right now. She's in leadership roles, shaping how AI is being crafted adopted and used.

Tim Odegard: She wouldn't say that but I can for her. So she's an example [00:05:00] of how it doesn't take flying a rocket ship to the stars to be an exceptional example of what it means to be a successful person with dyslexia. Hey, don't get me wrong, going to the stars in your own rocket ship is next level cool. Kudos, man.

Tim Odegard: Kudos. But many of us live our lives to make our own impact in different ways. And there is no one way to be successful at life and living life with dyslexia. So don't let some stereotypic idea about what it means to be gifted, have gifts, to think that you have to fit some mold that somebody else forced on us.

Tim Odegard: We all have our own unique set of skills that we'll cultivate based on our interests and our desires and where our lives takes us. So I want to introduce you to one highly successful exceptional dyslexic. Dr. Sarah Wood.

Tim Odegard: Transition [00:06:00] Music 

Tim Odegard: Well, Sarah, thank you for being here and agreeing to do this conversation with me. 

Sarah Wood: Yeah, of course. Thanks for the invite. 

Tim Odegard: You're really creating a space around assistive technology. What really motivated and drove you to really get curious about assistive technology in the first place? 

Sarah Wood: So the really brief answer is you can remediate a lot of the decoding, but my reading was not fluent in college. And So. There came to be a point where we had some cell biology class and we had these take home tests and it was use any scientific publication, you have 36 hours and you have to design the next study. And under those circumstances, that's when kind of the rubber met the road that there was no alternative other than trying to figure out how to use text to speech.

Sarah Wood: And at that point, it was like, okay, let me figure out how to do this. So that was like the moment that it absolutely became critical. There were times I tried it before, but as far as a crucible [00:07:00] moment, that was it. 

Tim Odegard: So it wasn't until you were an undergrad that you really felt that you just, mission critical, had to shift kind of how you engaged in reading for academics.

Sarah Wood: As far as being independently using synthesized speech, prior to that, a lot of the Shakespeare stuff I just got as audiobooks, which sounded better at the time, right? So I'm, this was in the 90s, synthetic voice wasn't what it was like, and some of these productions of Shakespeare are pretty good, that are recorded.

Tim Odegard: But you were listening to audio books through high school to try to get yourself through the readings. 

Sarah Wood: Inconsistently, I would say, but I also grew up in a privileged family where my parents were able to read to me. , you know, dyslexia runs in families, literacy runs in families. My parents were fortunate they were literate.

Tim Odegard: It wasn't that much earlier that I was going to high school, but The idea that I would have had in rural Arkansas, access to audio books for all the texts that I was having to read in my AP English classes that just they didn't have those. So I [00:08:00] didn't have that same option. 

Sarah Wood: Yeah, I was a semi rural student because , Santa Cruz Mounds is in the mounds of Silicon Valley, but it's in between Santa Cruz and Silicon Valley.

Sarah Wood: It got rural really quickly for some, but not as much for others. So it's this interesting kind of access point. 

Tim Odegard: One of the things that kind of distinguishes you and I, you had said the interventions you got had raised up your ability to do certain things. In your case, what type of intervention and when did that start and what did it look like for you?

Sarah Wood: So for some, fine and gross motor skills, that was early on and that was just some PT stuff. As far as academic reading, long story short, the public school itself was not responsive. My parents yanked me out of public school, put me in a private school for two years, second and third grade.

Sarah Wood: Thought, okay, maybe that's going to make the difference. That did not. And then starting right around that same time period, like, halfway through that private school, a separate dyslexia family. They had several different [00:09:00] identified generations of dyslexia. Their tutor became my tutor, and that was the Linda Mood Bell Program.

Tim Odegard: What dosage were you able to get?

Tim Odegard: So how often and for how long? 

Sarah Wood: I almost feel like I should have had the tutor on as the guest, because I don't want to misspeak on her behalf. 

Tim Odegard: You're not on the witness stand right now, and I'm not going to perjure you if you get the wrong answer when I cross examine you after putting the language therapist on the phone.

Sarah Wood: But it was at least twice a week, for sure. I want to say it was an hour and a half, but I'm pretty sure it was only an hour once you count travel time, because it took about 45 minutes each way to get to them. And so, again, that was the privilege of like, yes, I was rural, but my parents were able to afford to drive three hours.

Sarah Wood: But the school itself was unfortunately not helpful, to say the least. 

Tim Odegard: Which wasn't uncommon in the 90s, and sadly, from what I've seen by going around the schools, it's still not. I've got one more [00:10:00] curiosity about your initial answer, , I'm asking because I kind of know what my research has shown.

Tim Odegard: And so you said you were, in essence, still disfluent and weren't a rapid, automatic reader when you went to college. So, how did your spelling come online? 

Sarah Wood: Did my spelling come online is more the question. 

Tim Odegard: I know, right? 

Sarah Wood: So, I feel like in grad school, all of a sudden spell check became like reasonable, like my answers, I were actually like recommending things that were in the ballpark.

Tim Odegard: Yeah. 

Sarah Wood: So I feel like maybe like the last five years,, I have no idea. 

Tim Odegard: Right. We can increase the accuracy and the decoding concepts for the things that we directly teach and we get accurate to a certain extent, but the spelling often lags way behind and then also the reading fluency. So I was just curious if that was kind of what your experience was as well.

Sarah Wood: It must have been because I never remember passing a spelling test ever. My parents were very happy with a D in that subject. 

Tim Odegard: That makes two of us. Spelling is the bane of my existence. And [00:11:00] so that's curious. 

Sarah Wood: Yeah, I mean, actually, so Google, early on because of the way their algorithms were designed, that became my favorite spell checker in high school.

Tim Odegard: That makes perfect sense, actually, because I couldn't even get close enough to where I could use spellcheck. And then spellcheck wasn't looking for contextual aspects around the syntax to see if there's other reasons to believe that, oh, he really means this. I found spellcheck to be helpful, but as one of my buddies who's still a really close friend, would always say is, Dude, the problem is you keep spelling words to make them real words.

Tim Odegard: What advances in the assistive technology around, let's just say spelling now, do you think were kind of crucial from your experience that you think has better advantaged those of us in this community? 

Sarah Wood: I think that contextual AI knowledge of simply where the different ways to spell where. 

Tim Odegard: Oh you hate where too.

Tim Odegard: Like I say where and were and was. I just hate those. It's like the triumvirate of evil [00:12:00] W's. 

Sarah Wood: Right. Right. Right. Yeah, and just contextually, too, I, I'm able to now write a paragraph and I don't have to go in necessarily word by word and kind of just take that and be like, okay, here's what I think you meant.

Tim Odegard: Which is nice because often when you're working in a writing class or something to say, just get your thoughts on paper. It's like, if I only could, if you didn't realize how much effort I have to go into because of the basic mechanic of spelling, is getting in the way of me just saying what is in my head.

Tim Odegard: Do you run into the issue where I have five different words that I'd prefer to use as my first choice when I'm writing something, but I have to default to the very simple one that's not nearly as nuanced. Do you find yourself having to, in the sake of when you used to have to try to spell things, like, let's say on a in class, Writing assignment.

Tim Odegard: Did you find that experience? 

Sarah Wood: Oh, yeah. No, absolutely Still on medical form sometimes, you know, like the simple choice like fill in the blank [00:13:00] Like, okay, I can use these medical nuance terms, but no, I can't spell them in this, there's not an additional copy and it's too much asshole to ask. So I'll just use an easy kind of not technical term, but gets the point across.

Tim Odegard: Yeah. And that's something I think other people take for granted too, is if you go to the doctor and you're trying to just fill out an intake form and you're trying to say what conditions you have or what medication you have, it turns into a spelling test for those of us in our community. And you don't want to feel a bunch of shame and guilt because you just can't even fill out the form that you're supposed to do to get into the door.

Sarah Wood: Well, and sometimes you're like, Oh, okay. Like this seems accessible. It's on a tablet, but no, a lot of the accessibility features are locked down on this tablet. And it's something that I didn't really think about. I mean, I suppose I could call them and ask ahead of time for accommodations, but I read well enough where I don't automatically think about that.

Tim Odegard: And that makes perfect sense. 

Sarah Wood: But the medical field in general, as they move towards these tablets, there's not a technical reason that they can't have more assistive technology just [00:14:00] automatically baked in for everyone. And this community would benefit tremendously from that. 

Tim Odegard: They probably don't even think about it from that perspective though.

Sarah Wood: Yeah, I know. And that's like part of the problem is we have all these workarounds, but we also have the technology now to just not have to use the workarounds. 

Tim Odegard: So what would be a technological solution to let's just say something as simple as going in, you're checking in with your primary care physician, or you're on that follow up with a specialist and they give you that tablet that's wrapped in that big rubber thing to keep you from dropping in and breaking it.

Tim Odegard: What technology is currently available that they could be porting into these types of applications? 

Sarah Wood: Yeah, so having the ability to do the contextual spelling, spell check. For sure. For the short answer or insert things here. And then as well, having the computer read aloud or the tablet read aloud. 

Tim Odegard: Which should be easy fixes.

Sarah Wood: Yeah. 

Tim Odegard: So you got into assistive technology initially at, out of necessity, but [00:15:00] then I know that I was at a research conference and I'd already met you because you had interviewed to be in my lab years ago, but you decided to go to the Florida Center for Reading Research and work with a different scholar.

Tim Odegard: I was bummed about that. But I remember one of the research scientists saying, Hey, we've got this new student and And the real buzz there was they were talking about assistive technology, but nobody had done any research on it. So you actually thought differently. So what were you kind of interested from a research perspective when you were first transitioning into graduate school?

Sarah Wood: Yeah. So when I was still doing some cell biology and I have a biology undergrad background, they're like, Oh, here, try this. Try assistive technology. Try text to speech. They were prescribing that as if it was a. intervention and treatment, but there wasn't any guidance as far as how and when to use it.

Sarah Wood: And I kept looking and I was like, well, I guess we need this as a whole. This is a gap. 

Tim Odegard: Yeah. And I find that from my perspective, I've often [00:16:00] found those gaps in the research that seemed obvious to me, but nobody had thought to study or didn't think was meritorious of being studied. So, then what did you do kind of to try to help fill this gap?

Sarah Wood: Yeah. So, that led me to try and gather anything I could possibly find. And an older grad student at the time was like, Hey, you know, Sarah, you just turned this into a meta analysis. And I was like, Oh, I love numbers. Sure. Let's, you know, let's do that. Let's actually look and see if there is an effect size.

Sarah Wood: And there was. And we found it. 

Tim Odegard: First, what is a meta analysis? And what is it good to do? 

Sarah Wood: Yeah, so a meta analysis is a quantitative summary, if you will, of a whole bunch of studies to see if something, intervention, a different treatment path, really makes a difference. 

Tim Odegard: So then, what was the literature and what was the specific topic?

Tim Odegard: that you chose to do, and what was your outcome measure that you found was most often being looked at in these studies? 

Sarah Wood: The outcome measure was reading [00:17:00] comprehension. 

Tim Odegard: So you were looking at reading comprehension as your outcome measure, and then what types of assistive technologies were people using in these studies that you were able to find in your meta analytic review?

Sarah Wood: Yeah, so we really zoomed in just on the text to speech. Audio paired with a visual representation 

Tim Odegard: So, real quick then, so you initially looked at using synthetic speech generated by a computer, so going from text to speech, with them being able to see the printed text in front of them in some form. And when you looked at that, did you find that It improved the overall reading comprehension when you had the text to speech happening relative to when the text to speech wasn't happening.

Sarah Wood: Yeah, the scores did improve with the text to speech usage on the, on an individual level. There's a lot of heterogeneity within those results, like variation that wasn't accounted for necessarily by [00:18:00] the moderators identified. Like, age, I would have thought, would have been a significant moderator. 

Tim Odegard: So you identified age, what of the moderators did you code for?

Sarah Wood: What we coded for and what actually ended up being significant was such a, kind of a disappointment if you will, because age didn't really play as big of a role as I thought. Same thing with delivery method, whether it was recorded human speech or synthetic text to speech. The goal of the study, whether it was a intervention study or just kind of a compensatory Canvas tool help, and then what was read aloud.

Sarah Wood: Sometimes the entire test was read aloud. Sometimes just the passages were, but the questions weren't, and whether or not the material was on grade level reading or was kind of closer to their, their actual reading level. 

Tim Odegard: Gottcha. And did any of those moderators come out as being reliable predictors?

Sarah Wood: Not as much as you would think. 

Tim Odegard: I personally wouldn't have thought many of those would have mattered, depending [00:19:00] upon how the text in it was structured, for a couple of reasons from a developmental standpoint. 

Sarah Wood: Okay. Yeah. Go for it. 

Tim Odegard: Well, if we think of Dave Bjorklund's research, for example, on the usage of things, if you were to teach younger children to use something, we know that, let's just say from the memory Young children often don't think to automatically engage in a rehearsal mechanism to preserve or to help them in code.

Tim Odegard: But if you tell them to use it, they benefit it and they benefit just as much as the older children. So if we think about the old utilization kind of literature from cognitive development from the late 80s early 90s that if this is a in essence kind of an outsourced part of your cognition and you're being told or you're actually having to utilize it, The younger children aren't having to then themselves know to use this strategy, they're being told.

Tim Odegard: What would be the mechanism for age to matter? If it would be a utilization and you've actually structured it where you tell them to, or you have them use it to begin with, then there would be no age advantage when we typically see age advantages [00:20:00] in using kind of cognitive strategies in the established literature.

Tim Odegard: So what was your, what would, why would you might have thought that age would have mattered? 

Sarah Wood: It's just in the sense of, once you hit the, really, you're expected to, to read, to learn. You're not a, you know, you're not reading for the sake of getting your reading skills. You're really in the science and the history and just focused on getting the knowledge.

Tim Odegard: So a second reason that would kind of be another motivation to think that age wouldn't matter is, is as we've often learned from the research of Kate Kane and Laurie Cutting, we know in the earlier grades, they're really word reading tests, just as much as reading comprehension test. So then for that matter, it's not so much the reading to learn, right?

Tim Odegard: As much as it is in your earlier grades, you're actually getting access to the words and The heavy reliance on word reading to actually measure these is so much when you get to the older grades, all of a sudden your more complex kind of language skills are going to come on and be more important, [00:21:00] which is one reason why we often see the word reading measures not predicting as reliably you're reading comprehension after that kind of developmental point in time.

Tim Odegard: That's at least what I might have hypothesized. It's easy after the fact when you know the results, but as I started this, nobody really was thinking. That this was an important topic, but you did so extreme gratitude and kudos to you for just sticking with what you believed in. One of the things that I've noticed about you since I've known you is, is that you really are persistent and you don't give up on an idea.

Tim Odegard: And I really have always thought that was a great quality of yours. Are takeaways that you think are really critical? 

Sarah Wood: Yeah. One of the major takeaways is, like you mentioned strategies, how and when to use these tools. Now that we know they're effective, hopefully we're going to get more research soon. 

Sarah Wood: So I would agree wholeheartedly with you.

Tim Odegard: What kind of areas do you think are either being explored? Are you possibly exploring? What do you [00:22:00] think are some on the horizons in this area? 

Sarah Wood: For sure, the ability to. I don't see how students are utilizing this across the classroom and at home. We know home literacy, home reading activities are critical to reinforce the school learning.

Sarah Wood: There are programs that are able to kind of bounce between school and home and pick books up for fun. I mean, Harry Potter was the first book series I ever actually listened to and finished completely. But the whole idea of reading for fun, I didn't get any of that. 

Tim Odegard: Do you read for fun currently? 

Sarah Wood: I, so I have a ton of audiobooks, actually, now.

Sarah Wood: Um, because the last thing I want to do is look at a computer screen on my off time. 

Tim Odegard: That would be me too. I will listen to audiobooks, and so if I actually reprint on my off time, then it must mean that I care about that a lot. I'm curious, I've had the great privilege of being able to go into some of our oldest LD schools [00:23:00] and some of our largest LD schools.

Tim Odegard: For K through 12, the Shelton School was one that I remember, and it was about, it might have been about 18 years ago, I was there doing a teacher training, and I was able to also look at some of the new things they were adding in, and every single textbook at the Shelton School at that time was on an iPad.

Tim Odegard: And that was your text. There were no paper text. And that meant that every single iPad came with headphones. And that every single iPad went home with the students. They had created a ecosystem in which printed text were not the default. It was the eText that was the default with assistive technology built into it.

Tim Odegard: And that if you needed print, that would have been the accommodation. Like if you needed a actual physical printed copy, that would have been considered the accommodation you would have been given. What would be your ideas about implementations in more public school settings or in other settings that aren't these highly prestigious, expensive private LD schools?

Sarah Wood: I mean, I think the pandemic pushed that default by [00:24:00] necessity for different reasons. My godson was given a Chromebook starting in kindergarten, and so the kind of the default was he had a lot of stuff on the computer. 

Tim Odegard: Do you know of anybody who's doing any kind of monitoring studies, observational studies, looking at now as students are now digital natives and that we have these technologies that you and I didn't have growing up were embedded in them?

Tim Odegard: are the technology. I mean, Microsoft's immersive reading, for example, just strikes me. Um, my NIH postdoc, part of what I wrote into to do in year three, which I never got to, because I got my first assistant professorship before I finished out, was to actually use what's now embedded immersive technology, was borrowing from the attention mechanisms and feature binding to code certain orthographic patterns with different colors so that your default system would start to automatically recognize these as.

Tim Odegard: Orthographic units because your attentional systems and visual perception systems work so automatically in that way. [00:25:00] That's baked into Immersive Reader now in Microsoft. You can color code different patterns. Is anybody, to your knowledge, looking at what it might mean to be a digital native and have these baked in and what it might mean for our community?

Sarah Wood: I've heard people working on it, where they are in the process, Microsoft I'm not exactly sure and I can't speak for them. This is an interesting part because it crosses universities, it crosses companies with proprietary data and then it crosses some schools. 

Sarah Wood: What do you think the perceived use is? Are you finding people wanting to use assistive technology more?

Sarah Wood: So it's this interesting dichotomy, right, because there's some people, there's some kids who are absolutely insistent and when they can't use text to speech on something they're like, where is it? Why isn't it here? It's on all these other things I use. And then there's still other parents I met who are like, Oh, my kid wants to talk to you, but they're shy and I'm like, okay, and you go and talk [00:26:00] to them.

Sarah Wood: And there's a real stigma against using assistive technology. And so both are true simultaneously. 

Tim Odegard: What are aspects of stigma? 

Sarah Wood: Yeah, so some of the stigma I've seen is, you know, the kid doesn't want to be the only computer in a sea of books. That being said, I feel like that's more of a dated perspective.

Sarah Wood: Post pandemic, that's much less the case. The second one is they don't want to be the only headphones, right? So if there's a no headphone policy in the classroom with all these devices, then that automatically, boom, stigmatizes the text to speech user. 

Sarah Wood: So two simple structural changes for educators out there would be.

Sarah Wood: You often now have entire schools with just certain Chromebooks really adopted widely or other pieces of technology. And so, that seems like you're saying is like you're not being called out as being special or different by that. You're not being othered. But then, yeah, the headphone use, um, that could [00:27:00] be something that could be a preference to where if they want to use headphones, they could.

Tim Odegard: So that seems like two simple things that could happen, is if you're already in this technological sea, and you could have everybody have the option of headphones, are there other things that you're noticing? 

Sarah Wood: Yeah, so slightly harder ones are just making educators choosing programs that have embedded supports.

Sarah Wood: That being said now, I mean, Microsoft Word has a Google, there's a lot of plugins for Google that provide text to speech. But just, you know, kind of looking to see, okay, I'm going to adopt this for the classroom. Can all my students use it? 

Tim Odegard: Yeah, I mean, I think that's the one thing which is if we think about it as just we live in this world that is different than what any of us grew up in and it's shifting and morphing.

Tim Odegard: I don't think there'll be a day anytime in any of our lifetimes where we won't have to learn how to read, write, and spell. But, How we access and do that has changed dramatically just in the last five years. Grammarly was a [00:28:00] lifesaver for me. Before Grammarly, what I would do is I would listen to what I had written, because when I would read it, I wouldn't be able to notice, if you will, what the misspelled words were, or what even the missing words were.

Tim Odegard: I would, I would leave out words because I would fill them in automatically as I was reading. So I would listen to what I said, but once I got Grammarly, I don't find myself using what I needed this text to speech for anymore. So I do think that'll be an interesting opportunity to lean into as a society is to ask that question you just did, which is what I think I heard you say is, Why can't all students have access to these advances and these adaptations that we've created for ourselves?

Sarah Wood: Right. And, uh, you know, when I was in school, and actually in college too, I was told, well, you're not going to have access to this technology later. You're not going to have access to have your scientific papers read aloud. I, clearly that wasn't, that never came to pass because the access only increased as I went to grad school.

Tim Odegard: [00:29:00] You and I have shared things more personally about things that people have maybe with good intentions are not said that were really callous if you were to think about it from our perspective over the years. But, I'm sure that people who are listening have had more than once someone say that they wouldn't be able to do something because one of our perceived limitations and I think that you and I have both proven those people wrong by our ability to do what we've been able to do so and I agree that I've only found the access to technology Increasing and being more accessible at our fingertips.

Tim Odegard: So I would say that that's something else. 

Sarah Wood: So that stigma and fear, I guess, because I've had kids be like, well, but my mom says I won't be able to use it somewhere else. That's less and less true. Granted, there are times, you know, if you're studying in the middle of the woods and you only have a book and you're out of computer juice and you don't have a solar power charger, like, yeah.

Sarah Wood: You're not gonna be able to use text to speech. 

Tim Odegard: So that's true So what our perceptions are in the moment about what will be available. We're [00:30:00] not fortune tellers. We're not Nostradamus We're not gonna predict the future. So that's just you're that's just what you think dude, just to quote the dude, man That's just like your opinion, man I've had a lot of that in my life where people have tried to limit what they thought would be possible And it's like, man, I don't view those same limitations, man.

Tim Odegard: I just don't, I can't get on board with that. Like, I, I don't know that there's this thing that I often like to say, which is the rules of the game are set until you get to make the rules. So I do think we also limit the potential to think that what if they grow up and they make the rules? What if they're the ones that are designing the assistive technology?

Tim Odegard: What if they're the ones that are informing the policy that determines what happens? How limiting it is to think that. We wouldn't grow up to do whatever we wanted to do. 

Sarah Wood: We're actually starting to see that. Like, I'm starting to see younger dyslexia researchers say, you know, yes, I grew up with text to speech.

Sarah Wood: I use it every day. This, this, and this. And, and they're just not taking the [00:31:00] same crappy pronunciation that I'm used to, for example. Mm hmm. And that's completely fixable. With the way that we're able to, to mark up the computer words now, Yeah, the pronunciation problems can be fixed. We just have to adopt SSML, Synthetic Speech Markup Language.

Tim Odegard: So what is that? I don't know about that. So help educate me. 

Sarah Wood: Yeah. So Synthetic Speech Markup Language is simply a way to ensure words are pronounced as authors intend them to be pronounced. And there are products and companies that are using it. The World Wide Web Consortium has a working group related to it.

Sarah Wood: But I, I think like the main message for this podcast is simply the younger generation that grew up with these technologies is saying, Hey, no, like we want the standards to be this. We want the standards to be good and let's make that happen. 

Tim Odegard: What are other aspects of stigma that you see [00:32:00] emerging when you see the Um, individuals from our community trying to use or not use assistive technology.

Sarah Wood: Some of the SIGMA involves strategies too. The, we haven't had good instruction yet on how to incorporate text to speech in as part of a reading comprehension strategy and that any grad students out there, that would be a great area of work, how and when to use it as part of your kind of metacognition.

Tim Odegard: Well, let's just step back then. So, help me understand why or when you would use it. Because it seems like to me, one of the conditions for that type of study would be allowing them choice to use it, right? Typically what we do in these types of studies is allowing them to adopt it or not. And that might be compared back to an ever present type of a form.

Tim Odegard: It seems to me that if we wanted to think about the use of technology even in that, if no one's done this type of research, a first pass obviously, as you know [00:33:00] as a trained research scientist, is the simplest design, and then you go after the nuances later. Are you aware of any simple studies looking at the use versus not use of this?

Sarah Wood: Right. 

Tim Odegard: I mean, you did the meta and you found that it's effective. So it seems like it's generally effective. So are we at the stage now with the research that we need to move into more nuances of that? Or would it just seem to be that We know already that they just, we shouldn't use it defaultly or when we want to use it.

Tim Odegard: Is there something that you think might signal in the text that we should do it? Some kind of a monitoring or would there be also an individual difference there about a person's ability to be self aware and meta analyzed to begin with? 

Sarah Wood: That's exactly what I'm thinking. And also knowing when everything about comprehension monitoring in general.

Sarah Wood: When do I reread? When do I slow it down? What's my purpose for reading? 

Tim Odegard: Right. So if we think about kind of established literature here, if we were to think about the simple view, and kind of the reading profiles that come out of that, me having been more of [00:34:00] a, even though the IQ test didn't show it, I grew up to be more of your pure dyslexic.

Tim Odegard: Because my language comprehension, my inference ability, that's all been demonstrated to be on one test 95th percentile that I took to get into grad school. That's not a question anymore. It was when I was in third grade, but. I don't have DLD. I don't have a developmental language disorder. My issue isn't in more of those comprehension skills.

Tim Odegard: Do you think that a person who already struggles with comprehending in oral language might have struggles to know when they need to switch to using a print to speech because they may not feel a clunk? In what they're able to do, which those like you and I do, right? I, I won't speak for you. So I feel a clunk when I moved to print because what I'm able to understand is demonstrably different.

Tim Odegard: And I'm self aware of it. You're shaking your head, but for the audience, can you explain kind of what your experience is? 

Sarah Wood: Yeah, well, and my fluency. It's [00:35:00] just, it's, it's like my speed. All of a sudden I'm going verbally 80 miles an hour and then reading independently it's like 40, 30 miles an hour. 

Tim Odegard: Yeah, and that's interesting because that might be an individual difference as well.

Tim Odegard: Like, I know when I have to slow down and I choose to slow down and I choose to get really purposeful with my reading to make sure. I have to read every single word on the page. So the idea of skimming or skipping words that I was hearing from my buddies that they were doing, it's like, what are you talking about?

Tim Odegard: My first person experience was never of that. I just saw you shake your head. So what's your experience when you're reading as well, when it comes to that? 

Sarah Wood: For skimming, like, either I'm able to read the whole thing and understand it or I'm not. And so for skimming, I've literally adopted, like, read the title, read the bold words, read the first paragraph and last paragraph of each section.

Sarah Wood: I mean, what, you know, in college when you're trying to cram. So it's more about really being mindful of choosing what to [00:36:00] read and what not to read rather than a true skimming situation. 

Tim Odegard: Yeah. And I just never skimmed. I just read everything exhaustively. Just every single word. , 

Sarah Wood: I tried, but I, you know, just got exhausted and then I had advisors be like, well, you need eight hours of sleep.

Sarah Wood: So like, you're going to have to figure out something different. 

Tim Odegard: Oh, well, you know, that's just that, that was your problem. You slept. I never slept. So your, your health and wellbeing obviously was your main limitation there. You actually wanted to have a life outside of just reading and the ability to actually sleep, which.

Tim Odegard: I didn't get hung up on those little issues. So my thought is, is that one easier way of thinking about this would be to look at some very common profiles that we often already look to in the literature, which is if you come with more expansive language comprehension issues. Um, your ability to know when to use the print to speech wouldn't be as great.

Tim Odegard: And also it might [00:37:00] suggest that just having the, the print to speech wouldn't be as advantageous for you in the first place because you might need extra scaffolds and supports to help you glean meaning from it in the first place. I see you shaking your head. So what are your thoughts about that? 

Sarah Wood: Dude, I don't remember the nuances of some of these papers, but that, that was the take for some of the older studies that have been done.

Sarah Wood: But also sometimes built in vocabulary supports that come along with these wraparound text to speech tools. Some of them have embedded text to speech, but they also have vocabulary, some morphology, other type of instructional supplements. 

Tim Odegard: Now, we've been talking a lot about text to speech, but I think of Art Graeser's long line of research with auto tutor and technology using AI avatar and some of his recent work with adult individuals who were at least able to read, he said, at the third grade level, which we know that in our community, some of us can read at a third grade level, but we still have dyslexia.

Tim Odegard: So do you think that there would be a [00:38:00] avenue to build in some interaction? That would allow, especially for people who needed more comprehension support, to actually engage in a conversation that you and I clearly wouldn't have defaultly needed, may have benefited from, just to make sure that they were understanding what was being, that they were trying to read along with and being read aloud.

Tim Odegard: Would that might be beneficial as one tech solution to actually engage in that AI based kind of learning about the content. 

Sarah Wood: Absolutely. I think it would. And that's an exciting area that I think is ripe for exploration. 

Tim Odegard: Yeah, I would. I mean, and ART's been doing it, but it seems like what, , University of Memphis is currently doing is just a big adult study that they're running, and I don't see it porting into the elementary, middle, and high school, right now as far as what they're researching, but it does seem like an important way to do it.

Tim Odegard: ART highlighted that with, let's say, a common technology chat GTP, the turn taking in the discourse, so going back and forth in a conversation, it seems to lose [00:39:00] its way. And I find, as an expert in an area, I can definitely get it to go further and further and redirect it, but that's because I already know more than ChatGTP does.

Tim Odegard: So, I think that the use of those current modalities of the large language models that ChatGTP, at least, is using seems limited in its ability to do it without something more explicitly built for it, like ART's AutoTutor. What do you think are some of the promising things that could be used by our community for things that are freely available now?

Tim Odegard: Like Chat GTP 

Sarah Wood: I think prompting like small little bits, assume you're an expert in the following, even though you're an AI model, and then it kind of primes the pumps on the subject and just like many, many steps, and I'm able to get it to do more of what I want, but as far as easier tools. Texthelp has a lot of tools for, like, Equation Editor is great for kind of the math and the STEM parts of things.

Tim Odegard: [00:40:00] And what would it do? What does Equation Editor do? 

Sarah Wood: So, it crosses over domains. It enables you to kind of do the math succinct and kind of more naturalistic way, but also have all of my language supports built in. 

Tim Odegard: With this equation editor, when I think of equations, I think about the symbolic representation that we use in, let's say, calculus or linear algebra, those types of things, right?

Tim Odegard: Is it also being used in, for more word reading problems, which we know from Doug and Lynn Fuchs those of us at our community often struggle with because of our word reading problems. So it would actually be a great asset for potentially use for those word reading problems. 

Sarah Wood: Yeah, absolutely. Yes. For those reading, and if reading the math formula is allowed, like a lot of text to speech gets stumped at, historically, at math problems.

Tim Odegard: Because they're not built on mathematic kind of syntactic structure. They're kind of built more on discourse or other types of genres [00:41:00] of writing. Thank you. 

Sarah Wood: Again, with the advances in AI, that's changing rapidly. 

Tim Odegard: So, I often use Chat GTP when I'm reading now, and I run to a word that I think I know what it is, or like a concept that I think I know what it is, but I could ask Chat GTP to say, Hey, I'm reading this article about this, and I just ran into this.

Tim Odegard: What is it? What can you tell me about it? And then it kind of explains. I know that Chat GTP can confabulate, but I find at that level of prompting, it doesn't make stuff up, it kind of just gives you a nice little succinct summary, and now you can just have that read to you in a fairly natural sounding voice that would kind of explain a concept, and I like it a little bit more than just like going to the dictionary because For me, at least, it gives me a little bit more context.

Tim Odegard: It's a little more conversational opposed to a one to two sentence sparse definition that I could go to and get off the Oxford [00:42:00] English Dictionary online or do a right click and say look up off of an iPad or an iPhone. Do you find that helpful? Do you find others using something like that to try to help augment as they're coming into and learning an academic area?

Sarah Wood: Yeah. So I actually, I do that myself and then I tell it, Hey, can you cite some seminal papers in this area? 

Tim Odegard: Mm hmm. 

Sarah Wood: Mind you, the area has to be older, right? Cause I forget what it scraped the web in like 2002, 2001 for that, depending on the language model you use. So it, it can't get like the very latest emerging fields, but for some of the classic, it will just pull those papers right up.

Tim Odegard: That is definitely what I've found as I've been immersing myself in trying to understand the, the strengths and weaknesses of the freely available web version of Chat GTP. When I use that, I do ask it for sometimes to give me some seminal papers that I can go and look up in that. And I was just building out some new [00:43:00] training around the genetic pieces.

Tim Odegard: And I'm aware of it. I've been trained in it. I just don't know what the current stuff in the literature is. It was pointed to older stuff, but it was nice to get some summations from the Chat GTP, just like a textbook would be. What's your experience been with trying to, to work with the limitations that we know for Chat GTP to integrate that into your daily workflow?

Sarah Wood: Yeah. So for the daily workflow with the really most recent emerging papers, it's not. It's not very helpful, but for the trying to branch out and make a collaboration with a speech language pathologist to really kind of get myself up to speed with some of the basics in that field, it's great. 

Tim Odegard: Really is.

Tim Odegard: I mean, because it's generally accurate. You know, they, they talk about, I mean, it's funny, I've heard some of the people that have been into science, they talk about the fuzzy logic and I helped to develop a theory called fuzzy trace theory. It's one of the theories of cognitive development. [00:44:00] I'm not named on it, I just helped to test some of the hypothesis and work with Chuck Brainard and Valerie Reyna

Tim Odegard: and that's what they're talking about is, is in human thinking, and human memory and decision making, we often rely on fuzzy representations and we come up with things that fit the gist that are not 100 percent accurate because humans confabulate all the time and our ability to monitor when we're confabulating versus not is not something we do very well.

Tim Odegard: But in spite of that, you're saying that if it's in a general area, a well established area, you can kind of use it to bone up on what's being needed. 

Sarah Wood: They just know the basic vocabulary, the basic concepts, the basic theories. And I also insist that it gives me those seminal papers as references. You know, so it will summarize, okay, here's the five seminal papers and here's my summary of them.

Sarah Wood: And then I can go to those papers and read it. It lines up pretty well. Not all the time. 

Tim Odegard: Yeah. I found that it lines up well most of the time, but not all the time. And I think that pretty much [00:45:00] mirrors human cognition. Yeah. Yeah. is that the architecture of a large language model is not the neuronal architecture of our brains.

Tim Odegard: They're built off of different kind of, um, interpennings. They don't have neurotransmitters modulating the signaling of populations that are code signaling something. That's not how these large language models are working. But the end result looks very human when you look at where it does mess up. But I've, I've found the same thing that it seems to actually be pretty good at hitting the gist of what we've learned about a topic.

Tim Odegard: And I'm using it for a very similar reason. What other hacks have you found along the way to help you with kind of navigating the world when it comes to assistive technology? 

Sarah Wood: So a lot of dictating to my phone too, even like, so we'd mentioned medical forms earlier. Sometimes I'll ask, Hey Siri. Blah, blah, blah.

Sarah Wood: And then Cyr will pop up the written spelling of what I just said. 

Tim Odegard: Mm hmm. 

Sarah Wood: So I don't know what category that is in, but that's a hack. 

Tim Odegard: That is a hack. So you're using the speech recognition. [00:46:00] To then transcribe what you're saying so that you can still the spelling. 

Sarah Wood: Exactly. 

Tim Odegard: My wondering then is about variations in accent around the country and pronunciation, especially of individual phonemes.

Tim Odegard: As I joke, you know, I don't get hung up if a child says pen when it's supposed to be P I N because I know that there's many regions of the country that don't distinguish between the eh and the ih, especially in short syllable words. What is the accuracy that you're achieving in your use when you do that with your spoken accent with those technologies?

Tim Odegard: Are you finding that it actually understands you and is able to transcribe that at the level that's accurate? 

Sarah Wood: So for me personally, it does. That being said, I literally grew up outside of Silicon Valley and I'm white American standard dialect. I feel like by default it was designed for me without them saying that they, the default was.

Sarah Wood: A white Silicon Valley person. 

Tim Odegard: Well, Generalized [00:47:00] American English is definitely the one that's been programmed into these technologies historically. So, do you know of any work that's trying to account for dialectal variations and accent variations? 

Sarah Wood: Oh, absolutely. Especially as companies become global and we're trying to expand into marketing markets.

Sarah Wood: Absolutely. And people are no longer considering English as, like, the end all, be all, which I think is great. 

Tim Odegard: And then, you happen to be in California, and I know that in California, as well as many other parts of the country, we're dealing with large populations of multilingual learners. So, We've been talking kind of from our perspective of being English proficient and coming from English speaking homes.

Tim Odegard: Is there any possibility of assistive technology coming online to aid with either dual emerging classrooms or people from other countries who are either first or second generation immigrants to our nation here in the United States? [00:48:00] who are learning English and it's not their home language. Are you seeing any promising trends coming up with those types of uses of assistive technology?

Sarah Wood: So there are. I, I actually recently started reviewing literature on using text to speech in second language settings. And that's early on, so I haven't really looked at it. I can't give you an effect size number on that yet. But the practice is definitely there, as well as translation. Having English to whatever other language.

Sarah Wood: Translation as well as the pronunciation and the synthetic speech, I think we have, shoot, I don't know how many synthetic speech languages we have, but it's at dozens and dozens, if not hundreds. 

Tim Odegard: And by synthetic speech, we're again, talking back that to where it's actually marking up the language to have more of a naturalistic pronunciation of the words.

Sarah Wood: More naturalistic pronunciation and hopefully accurate as well. 

Tim Odegard: Do you mean accurate in 

Sarah Wood: context? Lead versus lead. Green versus [00:49:00] red. 

Tim Odegard: That type of stuff. Gotcha. So pronunciation variation for that orthographic pattern for that spelling. Right. That's awesome. That's great. I mean, that'd be a huge step forward.

Tim Odegard: And that's what our brains don't do very well. That's what I would always do is I would put the wrong spelling of a word in there. If it was pronounced the same way, like you were saying the different spellings of where, for example, um, as well as, you know, Knowing which pronunciation to use of a word, what I'm seeing it in context.

Tim Odegard: Those were always, those were challenging for us. And I can imagine for a, a non native speaker of English, those would be a challenge. And I also think it'd be a challenge for an AI system to know how to do that itself. 

Sarah Wood: Right. So that combined with the ability for text publishers and things to explicitly write in under the hood and just say, Hey, by the way, this is lead.

Tim Odegard: Right. Right. Right. That's really cool. 

Sarah Wood: And that's also happening. That's really cool. 

Tim Odegard: As a cognitive scientist, it's just so intriguing because I'm well aware that we've studied to describe what human language use can [00:50:00] do, but that doesn't mean that that's how we can engineer and program a computer system or computer agent to do it.

Sarah Wood: The other thing, it's just, it's, these tools are nice because they can do the translation, the, the speech, and it's a patient non judgmental practice partner, right? A neutral party, which I think is really important. 

Tim Odegard: You just hit on something that I found very advantageous and why I chat with Chat GTP is to really explore what I know, um, in a safe, in almost like a safer space than talking to a human that comes with the human nature to judge.

Sarah Wood: So there's that judgment factor, but then there's also a time, like you don't want to use all your time with somebody on the simple things. You want to actually get to their expertise and their meat of stuff. 

Tim Odegard: I'd ask, asking if someone of their time is, uh, is a lot. I think that's a really savvy view to think that way.

Sarah Wood: Right, and you can also attempt, like, multiple, oh wait, is this kind of right? I think I have the statistical model, but I don't quite understand this [00:51:00] loading factor. You know, and so that, to get that nuance, like, try, it's almost a deliver a practice partner. I. But I use that term cautiously because I don't actually know if it's capable of the evaluation for it to be a deliberate practice, but you get feedback.

Tim Odegard: You do. And that, in essence, this is kind of like why in the auto tutor, Art's shown so long in his published research that that dialogue that you can have in a more constrained way with how they program their auto tutors actually results in better understanding and better learning when you're engaging with an AI.

Tim Odegard: He's concerned, again, about the turn taking to where Chat GTP seems to dissolve itself. And also, since you don't have it, kind of, with a set knowledge base, that it's like a bona fide expert that's kind of cleared away all the confabulations, the, the gist based inferences that aren't quite accurate. Then you can do that better.

Tim Odegard: It's really intriguing that we can do it. I do think it's a limitation that some have identified that ChatGTB can only go so far with this, but there's been other things [00:52:00] developed that can be domain experts and actually go further on a topic, um, with it. So I think that's intriguing. 

Sarah Wood: Yeah, the rate of change with those large language models, I, I think we're gonna have a different conversation in a year about this.

Tim Odegard: I've already seen it shifting just in the last, six to eight months that I've been actively using it. With that said, I want to now switch to writing though, because I find that I've been able to do something that was a real roadblock for me. It was, I got through brute force. The reading is something I just took care of.

Tim Odegard: The writing was something that it was always laborious for me. And my written output doesn't represent the amount of research that I've done, how much I've shared. And everything that I've done, nor would my grant production mirror that, and one of the huge roadblocks to that was the written expression.

Tim Odegard: How have you either achieved success for yourself when it comes to writing, and are you seeing avenues for AI to aid [00:53:00] those of us in this community with that? Because we call dyslexia read instability, which is a misnomer. So what solutions have you been successful with in being an adept writer who's published at this point?

Sarah Wood: Right, so the, so my personal solutions going through school are I went to the writing center a lot, and I feel like now, actually, I can do this with ChatGDP. I've done it before, um, where I'll have a sentence, and then somebody will give me feedback and be like, the, you know, subject, noun, verb agreement is off.

Sarah Wood: And I'll go through with like real life examples for trying to learn that grammar pattern. Now with ChatGDP, it can just pop out a hundred different sentences. Okay, here's where the verb agreement's off and here's where it's corrected. Mm hmm. So that level of learning is helpful. 

Tim Odegard: I integrate Grammarly now into my first pass.

Tim Odegard: And the way that I'll use Chat GTP sometimes is I always do my first draft of a paragraph, of a section, and then I will feed it in and [00:54:00] kind of, hey, I'm writing on this topic, this, that, and that. Here's what I've got so far. What do you think? And it'll give me a list of areas of strengths and weaknesses after Grammarly's gone through and made it syntactically or grammatically accurate and got rid of the spelling errors.

Tim Odegard: That's kind of how I've been using it in that I didn't take the feedback. And then I incorporated it and I go through and say, I could strengthen my argument here. I could do a little bit better writing. So I've been using, in essence, a, a two layer approach. How does that strike you? Or what, how does that jive with what you've been using?

Sarah Wood: Yeah. Literally people have been like, I can't understand your ideas under the fog of spelling and grammar that you've had. And so absolutely, like, there needs to be a spelling and grammar pass just to clean up and make it somewhat understandable. 

Tim Odegard: Yep. 

Sarah Wood: Then for the second level, kind of constrained by what I can put out publicly in databases, but for locally run on my personal computer language models.

Sarah Wood: Just having a second opinion on whether or [00:55:00] not it hit my main points. 

Tim Odegard: Have you found any avenues that you've been using to kind of help you to jumpstart your correspondences or how you're communicating and writing with others? 

Sarah Wood: Yeah, Microsoft Copilot, for example, sometimes I'll say like, Oh, I need an email.

Sarah Wood: That I'm not sure how to word, you know, word it right, but it needs to contain the following information. I'm unable to attend this event due to, due to, like, you know, can, can you please make sure it's diplomatic or like figure out some neutral diplomatic phrases? And then there's like a chat function and there's a compose, like write about.

Sarah Wood: And then I'll usually throw in an abstract or all the information I need for an email. And you can set the tone, professional, casual, enthusiastic, and formal. Yeah, and then even the length too, whether you want us to be short, medium, or long. 

Tim Odegard: I've been able to be a lot more responsive to emails and more efficient in my daily workflow and save time.

Tim Odegard: I get a high volume of those, so it's a real, [00:56:00] Time suck. And it's been very beneficial to have these assistive technology pieces to aid me with just managing the diluge of emails that I get, most of them from outside of my job, just people asking me questions from all over the place. And I don't know kind of where your, yours is coming from, but I find that to be a real, real help.

Tim Odegard: Time sync for me, especially. 

Sarah Wood: Well, and actually, now that you bring that up too, I don't know about you, but other very successful dyslexics, emails like the vein of a lot of people's existence become a time suck for writing 

Tim Odegard: and reading. Yeah, I have a friend who's a successful lawyer. He always has a assistant.

Tim Odegard: And the one assistant was just, I'm not going to like spend time to like write your emails, read your emails, and like proofread your emails. And he's like, well, that's your job. If you want this job, that's what I hired you for. And I don't have to explain why you need to do that. And like I said, it's kind of, it's fun to grow up and actually set the rules.

Tim Odegard: And so [00:57:00] one of the things that we often presume is, is that we won't be in positions where we would be able to allocate resources to actually create our own structure that allows us to thrive and do what we do and maximize what people really should be getting us to do. Because people really want me to do something other than respond to emails.

Tim Odegard: And I'm sure that people want you to do other things, which brings us to something we really haven't gotten to, which is what do you do in the world? 

Sarah Wood: So I took a very meandering path, but I always loved science and loved research. And so I'm an accessibility scientist for educational testing service, but speaking on behalf of myself individually for this podcast, 

Tim Odegard: Sarah, it's been delightful.

Tim Odegard: Thank you for taking time out to talk with me today. And I look forward to our next conversation. 

Sarah Wood: Yeah. No, absolutely. Enjoy. Thank you so much.

Sarah Wood: Outro Music [00:58:00]