Human Information Processing

From CS1635 Spring 2016
Jump to: navigation, search




The Model Human Processor Chapter 2 of The Psychology of Human-Computer Interaction. Card, Moran & Newell. Only Pages 24-76.

Reading Critiques

Michael Oles 18:57:14 2/3/2016

I think the view this passage took of looking at a person as a computer was very interesting. There are many parallels such as processors, short term memory, long term memory, input and output. This can be a very useful idea for making both the computer and user interact better with each other. I thought the idea of a half life information was very interesting and a way we can make sure humans retain the information from a computer. Also the similarity that one processor can other do one deliberate task at a time. Sometime though I think this chapter may have oversimplified the human performance model and not taken into account what makes humans so much more sophisticated than a computer like you said in an earlier lecture when talking about UI.

Tiffany Martrano 13:10:05 2/8/2016

For today's reading, we had to read about the human psychology and cognitive processes and how they parallel computers. I thought this article was interesting, because I'm currently in a cognitive psychology class and could see the parallels between what the article talked about and what I'm learning in my psychology class. It was interesting to see the parallels and themes that I've learned in psychology appear in this article and how it relates to how computers also function. I was able to see the parallels that were trying to be drawn and it made sense to me.

Alexandra Krongel 17:27:47 2/8/2016

I thought it was very useful to have a few new paradigms for user interaction. The heuristic one was very reminiscent to what we talked about in class early on, what with there being cultural differences in perception, and how you cannot design for everyone all the time. A lot of Schniederman's rules also apply to web design, which is not so surprising. It's really hard to critique my own work but I know the principle of consistency is something I have difficulty with. I always want to implement something new and cool looking, but without regard to what users are accustomed to clicking on in other interfaces any 'cool looks' would be negated by confusion, and those features are useless then. While I know it shouldn't be over-relied on, text tags can also be useless here.

John Phillips 19:24:13 2/8/2016

I really liked this reading, and think it was probably the most interesting thus far. As far as actually applying it to developing good interfaces, I think think it could have been summarized in a page. The most important parts in that regard were probably where it discussed to how long it takes for someone to recognize a change (how fast user interfaces should be), how fast movement needs to be to look fluid (target framerate for animations), and how fast people can move (speed of typing and other input methods). However, while the rest wasn't really necessary, I thought it was very interesting. Often people quote things like "something needs to load in under 400ms to appear instant" or "you need ##fps to appear fluid" which are simplifications for certain cases. I liked how this article showed the exact results from different studies that led to most of the rules of thumb like these you come across in development.

Charlotte Chen 19:37:06 2/8/2016

The article presents a very scientific and mathematical approach in explaining human interaction with systems or software. It started off by going into details of the reasoning behind a person’s perceptual, motor, and cognitive systems. Later, the article used our understanding of the 3 subsystems into explaining designs of machines and softwares. I find the examples to be very interesting and intuitive. For example, one very familiar example is calculating the durations of lags between animation frames in order to create a smooth movement based on human’s rate of perception. The power law is also an interesting subject, and the design of the keyboard is a good example used in explaining it. Interference with working and long-term memory is a topic that I think will be most relevant to for our class. Since users will find it really difficult to recall similar items, we need to be very careful in designing the layouts and choosing hint words for our applications. For example, users will find it really unpleasant experience interacting with our app if we put nav-tabs on the top in stead of the bottom, or use the word “delete” in stead of “remove”.

Jonathan Blinn 20:30:47 2/8/2016

This was not the reading I was expecting. This seemed to be focused more on reaction times and the times it takes for a person to be able to learn, correct, or sense different things. While some of it was interesting (such as the pen experiment), I'm having a hard time with relating this to a mobile interface course. I'm hoping tomorrow's lecture will help relate the two more. While most of the reading felt fairly unrelated to the course, there were a few connections that I was able to make. For example, when trying to design a good layout for a keyboard or some similar interface, it is important to know the most commonly used buttons/keys in order to arrange them in the optimal way. For example, as an English speaker, putting the Z, X, and Q keys as the "home keys", it would be harder to access the keys we use the most while typing on an actual keyboard.

Daniel Hui 21:16:34 2/8/2016

This was sort of a tough read to comprehend. I understand that the article is attempting to make and analogy between human physiology and the internals of a computer system, but the overall purpose was unclear to me. Maybe it was because of the lack of terminology I understood in thus article. For instance, the part on working memory got me really confused. I didn't quite understand the purpose or point about dividing your working memory into chunks. Overall seems like an interesting article, i just failed to understand the comparisons and concepts given.

Andrew Lucas 21:32:37 2/8/2016

I found this week's reading very interesting, not mainly for its applications to android interface design, the connections with which seem tenuous at best, but for its potential impact upon machine learning, my personal area of research. Machine learning models such as deep neural networks constitute attempts to align the way in which computers predict future events based on past experience with the way humans do it. The article's decomposition of the workings of the human brain into memories, processors, and principles, and its further decomposition into perceptual, motor, and cognitive systems provides a new perspective on the problem. The perceptual system would be the input, the motor system would be the output, and the cognitive system would be the predictive model.

Nate Patton 22:12:42 2/8/2016

An interesting read. Very long, and at some points dry but it did make me think multiple times. One thing was how the brain, as a Conceptual machine calculates the distance for everything. Even the little things like the distance the person's finger travels for a keystroke on a calculator.

Max Benson 23:11:57 2/8/2016

This strikes me as a strange way to approach human behavior, although I see how it could be useful in terms of timing when events occur in an interface, as well as how interface elements are laid out on the screen. For the most part though I think the author's strict quantification of the highly subjective acts of memory and perception don't in the end provide very much insight beyond serving as a baseline metric of how to present data. I think discussing the human brain in terms of a simplified computer system is a novel perspective, but it is generally much more useful to think about behavior on a qualitative, case by case basis rather than trying to catalog the vast multitude of discrete functions which make up human experience.

Matthew Reinhold 0:58:53 2/9/2016

I feel like the reading was useful in showing how making your app easy to use the first time is very important in design. The mind is efficient at remembering things and associating things, so having a good first experience with your app is equally important as having a really good idea.

Joshua Fisher 1:11:04 2/9/2016

I really enjoyed this article and its comparison of the human brain to a computer processor, but I felt that some of the calculations and equations given were difficult to understand. I feel that the general premise of the article is useful for interface design methodology. I also feel that the idea that this needs to be factored in when designing an app's UI is an important concept.

Alex LaFroscia 1:14:26 2/9/2016

I found it pretty difficult to determine how to apply the formulas provided to the design problems that we would experience in software development. I understand what the chapter is getting at -- if you can treat a human brain like a computer, and come up with formulas that model the speed and effectiveness of humans at particular tasks, we can use the formulas to tune our designs to ensure that our users experience what we want them to experience. However, I have a hard time imagine a designer working with these complex formulas rather than going by trial and error or getting feedback from users. It seems much more complicated to try to derive a formula that represents whatever action in the application you're hoping to perfect than it would be to just ask enough people that you're comfortable with the result.

Luke Kljucaric 1:18:11 2/9/2016

I thought the readings were very interesting. It's different to talk or think about the human mind as being a processor. Its not just any processor that can be modeled quite easily as explained by the reading. It seems to be multiple different architectures of processors all working as one making us who we are. The reading really goes into depth into each system the brain has and at some points the reading can seem dated (Based on the research at the time). Also, I thought some parts were kind of hard to follow like some of the equations in the human performance section. I also thought the section on memory was interesting. It almost impossible to believe that in a bunch of chemicals we some how are able to store events that happened to us in the past and can recall that based on certain stimuli. I thought the interference and searching sections were a good read as well. Some of the sections towards the end seem to be cut off so it can be hard to follow at times but overall a very interesting read on the human brain's architecture.

Jason Naughton 1:27:25 2/9/2016

By now, I'm so used to reading technical documentation. I was pleasantly surprised by this nearly-satirical approach: technical-izing the user. The user in traditional technical writing is generally a snot-filled sack of skin that needs to be hand-held through the details. This writing was deliberately strange and contrived. Still, it's an interesting idea: that the user is a complicated machine--just as much a part of the application to be worked. And although the models were physically inaccurate, the concepts, like "Name Matches" or "forgetting just-acquired information" have their basis in the real world.

Chris Finestone 1:32:32 2/9/2016

Interesting, but a bit of a long read. I think every user is different and trying to get interfaces down to such stringent methods for evaluation doesn't really model real world use.

Zane Hernandez 1:41:46 2/9/2016

I think this is a very stupid and archaic way of looking at things. People are not computers. There is a lot of effort put in to this nonsense and I can't understand why. I don't think you can apply engineering concepts to how people react to or feel about stimuli.

Casey Nispel 1:42:42 2/9/2016

While I’ve heard of the human processor model before, I’ve never actually used it or seen a detailed breakdown of how everything is calculated within it. I found it really interesting how every single aspect of the human mind can be broken down in to simpler components and grouped within either processing or storage areas of various areas of the human mind. I always wondered how accurate these calculations can be though, since the human mind is so inconsistent and so many variables can affect things like reaction time and recalling things from memory. While the model can help predict things like how an average person or an incredibly quick person would respond to certain situations, it still doesn’t compare to running a test with a real human and getting real human response times.

Dustin Chlystek 4:54:07 2/9/2016

I found the way he compared humans to computers interesting. I like how he used standard reaction times from humans as measurements. One in particular I found interesting was the part with the clicks. I like how he described that humans can only register click so fast, and how when 10 clicks play in a second most are heard, but if you up it to 15 or 30 then the clicks seems to run together because humans can only react so fast.

Yijia Cui 6:41:35 2/9/2016

Today’s reading brings up a really interesting idea that human mind is also an information-processing system. The Model Human Processor is defined as a set of memories and processors and a set of principles. And it can be divided into three subsystems: the perceptual system, the motor system, and the cognitive system. For each of those subsystem, they all haves their own memories and processors. The perceptual system represents the physical sensation collected from body sensors and integrated into inner world. As a result, the perceptual memories are used to hold physical stimulus. It is interesting to know that the perceptual memories are closely related to Working Memory. The motor System carries out motion translated from the results of the perceptual system and the cognitive system. The cognitive system serves two different levels: at a simpler level, the cognitive system is only the connection of the perceptual system and the motor system; but at a higher level, it would involve complicated processes with information just received and stored for long-term. Because of the complexity of the tasks carried out by the cognitive system, it has more complicated memories and processors. After giving a detailed introduction to those three subsystems, the author started to analyze examples of human performance for the relations to human-computer interaction and also the Model Human Processor. The performance is composed of perception, motor skill, learning and retrieval, and problem solving. All in all, this author provides more “scientific” ways to relate human to computers, which would be a good preference to design stuff that can be fitted into both human needs and also computer expectations.

Lmn33 6:43:45 2/9/2016

I found the eight golden rules in the reading to be mostly about basic things to do, but could see how someone could forget to put one of them in. The one I didnt understand was Offer error prevention and simple error handling. I was not entirely convinced about this one. To me the only true way to handle this is to have a team whose sole purpose is to wait for someone to have an error and then fix it. Allowing another user to be able to do something to me is too risky.

Sarah Dubnik 7:39:50 2/9/2016

I found this to be a very odd reading and somewhat discomforting to see the human mind broken down in this way. At first I could not understand the point of some of the data presented, as many of the numbers seemed to be estimates or based on very few data points; furthermore I did not see how this information could be useful. I understood later on that things like perception rate, for example, can affect the interaction with something like an app, so it made more sense later in the reading; however, I still think that much of what I read was unnecessarily detailed and less relevant than previous readings.

Nicholas Carone 7:47:11 2/9/2016

This article is an interesting take on the human perceptive abilities as shown through several different information processors. When the human body is viewed as an interconnected system(as it is) and similar metrics are applied to the performance of computers as humans, real data can be collected as to the abilities and capabilities of this 'human processor'. I enjoy how the researcher breaks human cognition into several different processors and test their interactions as well as their performance as a whole.

Clark Nicolas 8:39:35 2/9/2016

This reading was a bit too math-y for my liking, but I do think it's interesting that human performance has been so rigorously studied. I particularly enjoyed the section concerning long-term memory because of its relevance to learning new programming languages.

Ishvaraus Davis 8:40:21 2/9/2016

The reading today was very technical but it provided an extremely large amount of information. The author presented the human as a model that was very consistent and predictable, when we tend to think of humans as more sporadic and unpredictable. The insight provided into biological and chemical systems like memory and motion were very useful and I think they will be beneficial to use one day.

David Fioravanti 8:46:10 2/9/2016

This reading talked about the human brain and how it functions like a computer. The visual system takes in a large amount of data every second. However, that amount of data cannot be processed all at once because the brain has a limited capacity. The brain will filter data based on perceived importance given visual or audial cues.

Amukher14 9:01:12 2/9/2016

It was interesting to learn in depth about how the human mind works. The speed of the processing of the mind allows me as a developer to work around that information to know how fast a system needs to be in order to be considered responsive.