Human Information Processing
From CS 160 Fall 2008
Lecture on 10/6/2008
Readings
This is a classic text on low-level human behavior and modeling, but it is not an easy read. You do not need to remember all the equations. You should focus on architecture of the MHP, the "principles" listed, and on two of the quantitative laws: "Fitt's Law" and the "Power Law of Practice". Grab a cup of coffee and find a not-too-comfortable chair and enjoy the reading :-)
Discussion Questions
As the examples show, the MHP has mostly been used for low-level modeling of UIs. What aspects of game design could use these results?
Discussions
Please post your critiques/commments on the required readings below. To do that, first login by using your user name and password, then click the "edit" tab on the top part of this page (between the "discussion" page and the "history" page), New to wikis? Read the Wiki editing guide. . Hint - Please put a whole line == ~~~~ == (literally) at the beginning of your submitted critique, so the wiki system will index, sign and date your submission automatically.
Mike Kendall 19:30, 5 October 2008 (UTC)
The model human processor does seem to be applicable to game design just in a general sense of human computer interaction. For example, a designer should avoid putting information on the screen for too short of a period for it to be interpreted (I'm looking at you, skill point bonuses in Xenosaga 2). I don't quite see an application in the design process for the power law of practice though. In lots of games, the player will become better ad better at doing the tasks required of them, and this is part of the game. In some games, you get a reward for being faster, in others its rewarding enough to be able to get the task done fast.
In RPGs for example, when you first encounter a new enemy, you have to figure out how to beat him. as you fight this enemy more and more, you learn to beat him faster and faster. In most of these games, there is no reward for beating them faster other than saving you time, and that's enough... Then again, the power law of practice only really has to do with interaction, not design, so it seems hard to apply it to design...
Fitt's law, on the other hand, I don't see as applicable even in interaction... It seems more like a rule governing moving from point a to point b. I guess one consequence of fitt's law is that the destination point should be an area rather than a point to make it easier for the user to get to.
The discussion about memory seemed applicable to a game, especialy an RPG because oftentimes something from earlier in the game (by hours and hours) can have an effect on you much later... If you want something like that to make sense to the user, it will have to be estalished in their long term memory by making reference to it and related things repeatedly. Another way that I see this as appicable to RPGs is the recent trend of giving a small synopsis of what is going on when you load your game. This is re-establishing your working memory from the last time you were playing, which is pretty cool.
Perry Lee 22:18, 5 October 2008 (UTC)
I agree with Mike -- the MHP seems applicable to game design just in a general sense of HCI.
Of the listed principles, those that stand out to me are the memory-related ones (e.g., encoding specificity principle) and the uncertainty principle; they seem to play an important role in easing the learning curve of an application or a game. As was mentioned in previous articles (although for different reasons), it is important to provide cues so that people can more easily memorize/recall information (e.g., using an acronym). You also want to be careful not to flood people with too much information in too short of a time.
Another issue to be aware of is the uncertainty principle, which shows that decision time increases with uncertainty about judgment. In a game (and really in any task that involves memorizing/recalling), you want to draw upon as much of individuals' past experiences to minimize the uncertainty people may feel when using a new application or playing a new game. Familiarity is a good thing; if you can draw upon a similar task that people have done a lot in the past or currently, you will be able to take advantage of the power law of practice, which simply states that the time to do a task decreases with practice.
Kumar Garapaty 22:47, 5 October 2008 (UTC)
I found this article interesting because I have never heard of this combination of psychology and processor analysis to examine human computer interaction. I think this type of analysis is especially useful for more tangible interfaces rather than the game design in class because these equations might be more accurate/useful for those kind of interactions. I think it would be rather difficult to use the mathematical analysis they used in this article to examine the design of a certain software and even if they did, I would doubt it would be of any significant importance. For game design, you can just measure the usability by simply noticing how they use the program not by calculating how much reaction time it takes to perform a certain action. I think you would want to know this information for highly specific games that would depend on some kind of timing for memory and reaction in their design/interface. And even if you did need it, you could probably approximate all this information without doing any of these calculations. Perhaps this information might be necessary in highly complicated games, where rules and features of the game can be misinterpreted or forgotten. If it was a tangible interface we are discussing, then maybe the various speeds for using an object such as a mouse would be important to make sure the user does have the time to appropriately use the mouse. However, I think games along with other items that humans interact with can gain from the general concepts of MHP.
Stuart Bottom 00:23, 6 October 2008 (UTC)
In particular, two aspects of the MHP strike me as being especially applicable to games. For serious games, the model of memory described in the MHP seems particularly useful in determining what techniques can be used to maximize retention of key concepts. For example, increasing the number of associations to an object in Long-Term Memory increases the ease with which it can be recalled (cf. discussion of P3. Discrimination Principle on pages 40-41). Games that actively assist or encourage the player to increase the associations to the items they're learning will help the player to more effectively retain the material.
I would argue that Fitt's Law does in fact directly apply to certain HCI questions, with perhaps some slight modifications (maybe you'd need to add in a small constant to account for system-dependent uncertainties induced by things like the "stickiness" of a mousepad or variations in the accuracy of a laser mouse - in extreme cases, the computer itself may induce delays, such as if the user is asked to manipulate a 3D model that must be redrawn after the view is changed). Essentially, though, we are still talking about direct manipulation: the user interacts with a mouse or joystick, observes the motion of the pointer on the screen, and corrects the trajectory as necessary. The action is the same; it is merely the feedback system that is different. One could just as easily argue that you are required to move the mouse to a specific target on the mousepad. Replace this with a pointer on the screen (which is usually a higher-resolution feedback system) and it is still a question that Fitt's Law addresses.
Gary Wu 03:17, 6 October 2008 (UTC)
The perception section of the MHP article definitely applies to human computer interaction. Things can't be too far off the center of the screen that are important. If status alerts and messages are in the peripheral region of the screen, judging by the formula, there is a certain amount of time it takes our eyes to register new information. Thus, those messages would have to stay on screen longer or flash a bright color to grab the eye's attention. Another statement that I found particularly interesting was that movement is not continuous--but rather it consists of a series of discrete micromovements, each requiring some finite amount of time. I found the idea of the mnemonic device of using chunks in Working Memory to be a very sound theory. I can definitely attest to doing this in the real world all the time.
In serious games, it is definitely practical to predict human performance less the target audience be very small. Having a slowman, fastman, and middleman mode for the game could be for difficulty. This allows a broader range of users to feel comfortable with the product. Fit's law definitely applies to user interfaces. Designing a product that does not consider Fit's law would be a horrible product to use and have a very messy and unintuitive layout. I can now see why many software applications have like tools grouped together--to minimize the distance between the hand and the target. The Power Law of Practice is applicable to superusers of a particular product. I feel that hotkeys definitely fall into this category. It allows the user, who after a few tries, to become more productive and efficient. The learning slope of a product will definitely play a role in how the user integrates with the product.
Jacekmw 04:45, 6 October 2008 (UTC)
In general the idea of the Human Processor is an interesting one to me, though I have thought of this it was never quite to this level of detail. The applications to games in this case are numerous. Games, in a sense, can be split into categories, as some rely mostly on the Perceptual and Motor Processors, as in some twitch-and-shoot first person shooter games, while only those requiring strategy utilize the Cognitive Processor to its maximum capacity, to more than just the Discrimination and Encoding principles. However, most of the other principles and laws MHP describes can also be applied to games. Fitt's law is a clear game-interface-limiter, as limiting the speed users can move their hands is precisely what makes first-person-shooter games as difficult as they can be. The Power Law of Practice, on the other hand, reminds me of older games such as the Megaman series, infamous for the amount of times one would need to replay the stages in order to finally know strategies and have the muscle memory to beat them completely. The Uncertainty Principle is important in that many artificial intelligences of enemies utilize random elements, making the result states of your moves not entirely dependent on what you do. The Rationality Principle can likewise be used in enemy AI behavior in order to anticipate what you might likely be planning to do, while the Problem State Principle reminds me of mystery games such as Myst which rely on gathering information and items (which can be thought of as operators in this cast) to advance in the adventure.
JoshuaKwan 05:03, 6 October 2008 (UTC)
The MHP studies basically describe humans as if they were computers, with response times and fast-access memory capacities, and goes on to describe how to exploit our response times to provide the illusion of reality in computer graphic simulations using a bunch of hokey formulas similar to the things that computational economists use to predict whether the world is going to collapse, except that these are actually based on empirical models. Being able to model human response times to visual stimuli like computer screens is very useful for deciding all sorts of timing issues in a game, e.g. how long do you allow for a player to press a required key before the time expires and the player has to start again, or whether a certain stimulus can get the player's attention at all.
Xuexin Zhang 05:29, 6 October 2008 (UTC)
Similarly to other UIs, the basic level modeling of game interface design could use the result from MHP as it provides many valuable feedbacks to UI designers. I disagree with Mr. Garapaty's ideathat is difficult " to use the mathematical analysis they used in this article to examine the design." Personally I believe that the reaction time is a very important way to find out if the design seems logical to users and by keeping track of such information, UI designers could learn more about the usability of the program.
Jimmy Nguyen 06:01, 6 October 2008 (UTC)
This was a very interesting article on how the HCP thinks as a person to receive input and interpret what to do with the input through the perceptual and cognitive processors. The most interesting section I found was the Morse code section, how they integrated that into everything. In terms of serious games, I feel that it definitely applies to a lot of games, since a lot of games are based on how quickly you accomplish something, or memory based. Those are the two obvious ones, especially in terms of serious games in which the user is trying to learn how to do something, or trying to learn to do something faster.
Mohammed Ali 06:16, 6 October 2008 (UTC)
The three realms of The Model Human Processor has a lot to do with game design. All three interactive subsystems, namely the perceptual system, motor system, and the cognitive system all play a major part and collaborative nature in game design and execution. It can summed up in perception, cognition and reaction. The gamer sees whats going on storing the information in buffer memories while awaiting symbolic recognition of the events being depicted on the screen and then uses their cognitive part of their brains to make decisions and then finally their motor skills to react to the events.
Fitts' law can have a lot to do with games. This is because games serve as an entire user interface quite often. In a menu for example if the selections are to made with a pointer-like tool, then practicing the functions within Fitts' law is essential. For example, having menu options lined up across the edge will minimize, actually eliminate, the capability of overshooting. This will enable faster access. The icons, of course, should also be made larger adhering, again, to Fitts' law.
The Power Law of Practice is also very applicable to game design. As gamers learn how to access items, learn attack combinations, explore the control configurations, or whatever tasks they choose to do amongst the array that available, they will learn to execute those tasks faster the more times they do it. This principle, just like Fitts' law is quite intuitive. Game design heavily uses the Power Law of Practice.
Haosi Chen 06:40, 6 October 2008 (UTC)
An issue to be aware of in the article is the uncertainty principle, which shows that decision time increases with uncertainty about judgment. In a game, you want to draw upon as much of individuals' past experiences to minimize the uncertainty people may feel when using a new application or playing a new game. Familiarity is a good thing; if you can draw upon a similar task that people have done a lot in the past or currently, you will be able to take advantage of the power law of practice, which simply states that the time to do a task decreases with practice.
KevinFriedheim 06:41, 6 October 2008 (UTC)
The Model Human Processor (MHP) describes a mechanism that really covers a general base of how human interaction relates to games – but I think that game interaction is of a whole other class of its own in that certain rules/principles as described by the MHP. However, certain ideas that MHP presents, like exactly how long it takes a person to type a particular subset of keys, can be useful in modeling games. Perry Lee makes a good point about how uncertainty needs to be resolved as it could be crucial to game play (and possibly unbeknownst to the designer of the game). I’d certainly have to agree with the fact that the MHP brings up about relying on long-term memory for decision making and how it takes longer in the process. I can recall plenty of times while playing “poorly designed” games that I was stumped when I ran into an object or situation that relied on my remembered how I treated/reacted to the object/situation earlier in the game play storyline.
Vedran Pogacnik 06:50, 6 October 2008 (UTC)
Fitts Law, that analyzes the relationship between the time to make a movement versus the relative precision of the task required, seems too singly-dimensional. It seems to be based on “cycles” of human processing. The approach indicates that humans cannot chew gum and walk at the same time. Also, it doesn’t take into account the dexterity of different parts of the human body, specifically arms (with their limited flexibility) and shoulders. Also, what if the tasks involved the use of two hands? Does the law specify concurrent processing or is it sequential? I think that there is a larger and much more complex context in which we need to consider assumptions presented in Fitts Law, and hence Fitts Law by itself, is incomplete. To back up my argument, would Fitts Law still hold if the calculator (example 5, pg 56), was held in your hand but with the arm stretched above your head towards your back? I doubt it.
On the other hand, the Power Law of Practice seems sound, although I couldn’t find the part which accounts for asymptotic limits of the time required for each task (as indicated on figure 2.13, pg 59).
The principal aspect of game design that could benefit from the studies presented in the paper is the grouping of actions, that are required to play the game) by functionality or frequency of use, depending on the nature of the task.
The problem space principle reminds me on a common bug in many programs: how many seconds goes in a minute? One would say 60 immediately, but all of us know that that might not be true on Mars (either the measure of a second is different, or seconds to minute ratio, or minutes to hour, or hours per day, …).
Saliem Than 06:58, 6 October 2008 (UTC)
I found it interesting that the perception component of the model allows for the calculation for things like the illusion of motion, causality and such. The article noted that the cycle time of the perceptual processor setting the speed at which a user can attend to an auditory visual or auditory output. I thought that that was relevant in setting up feedback cues when designing games. In designing serious games, the notion of the discrimination principle might be useful as well. Assuring that the game has distinct objects throughout all the levels a user might progress through might allow the user to learn more. Lastly Fitt's Law and the Power Law of Practice might be applicable for tangible user interface designs and the games that make use of it.
Jeffrey Rosen 07:09, 6 October 2008 (UTC)
It is important to take into account human capabilities when designing an interface. While many of the things listed in the article are extremely specific and not too useful, for example the exact time it takes to memorize various numbers, some of them are extremely important. For example, choosing where to locate icons on the screen. According to Fitt's Law as discussed in the article, you can measure how long it will take a user to find various items based on their location. For instance, it is very easy to hit a menu bar at the top of the screen. If that menu bar was located in the middle, or some other location, it would be much harder to navigate to using a mouse. Corners are the best, because there are two edges. You can see this clearly representing in OS X. All of the corners are used for important UI elements.
Buda Chiou 07:13, 6 October 2008 (UTC)
The Model Human Processor may be only related to the interface design of a game. The interface, however, plays a very important role in game design. A successful game may not have good plot or complex operation, but it must have a user interface that makes users feel comfortable with. The reason is that user will see the game user interface prior to any other component of a game, so if the interface is bad, no one will continue to play the game, and Model Human Processor definitely can help the game designer to avoid this kind of problem because the more we know how human react to things, the more we how what user want.
Jordan Berk 07:21, 6 October 2008 (UTC)
Fitt's law applies very directly to HCI, including games. Think of the Sims, with its easy to navigate pie chart selection process, as opposed to some games which have seemingly endless lists of actions or options which take time to point to, especially if there's small text. The Power Law of Practice is also essential to factor in for gaming, since providing a challenge for a user is often a primary requirement for how fun/interesting a game is, including serious games. As the Power Law of Practice comes into to effect, and without gradually increasing game difficulty when it comes to the repeated actions governed by the law, the game will become too easy to play, and therefore not provide the necessary challenge, and by extension, fun.
Wenda Zhao 07:23, 6 October 2008 (UTC)
Model human processor has two major components: memories(long-term memories and short-term memories) and processors. The model shows how users are able to retrieve task-related info from both short-term and long-term memories in response to stimuli from different senses. We can conclude that people can perform complex tasks that involved in more than one sense at the time. So when we design the game, we do not have to worry about if we fed more than one sense at once will make users confused or annoyed since we know that humans are good at processing more than one sense at once and they can also handle info simultaneously from both long-term memories and short-term memories. The game play can incorporate with visual, auditory and touch system at the same time to increase the excitement of the game.
Kai Lin Huang 07:29, 6 October 2008 (UTC)
In the human processor model, perceptual processor is much slower than the cognitive processor and motor processor, so the following aspects of game design can use these results. Clearness of layout: Because perceptual processor is slower, game control buttons for reaction activity should have clearly visible direction and be as concise as possible. Players will spend more time in analyzing the display instead of enjoying it. The timing of a game: The easiest level of a reaction game must be something slower than the sum of processing time of the perceptual, cognitive and motor processors. More players can adapt to the game when it starts at a speed that’s slow enough. Graphic instead of word: Because human may memorize by encoding what they perceive into something abstract, a memory game for vocabulary can use picture-word combinations instead of purely letters-description, so the player can easily retrieve their memory of the word by accessing their image library and find the word that links to the specified image.
Alan McCreary 07:44, 6 October 2008 (UTC)
Fitts' Law seems particularly useful for game design. As Vedran pointed out, it is one-dimensional; but if the interface we're designing simply involves moving a cursor around and clicking (as with many computer games), Fitts' Law suggests that making important buttons large and relatively close to one another is a good idea.
On the topic of serious games, the Encoding Specificity Principle is important – it says that if we want the user to remember something (how to brush teeth, for example), we should make the “learning” and “applying” situations as similar as possible. With brushing teeth (as in the example we saw earlier in class), it means that instead of having the user learn by dragging a graphical toothbrush over a set of teeth on a computer screen, the designer should make it so the user learns through actual physical movement of a toothbrush (or toothbrush-like object).
Bing Wang 08:43, 6 October 2008 (UTC)
Many of the examples in the reading can be considered as one of the factors for game design. The reading starts out with explaining the basic components of the human processor and later on gave examples of what each of them do in real life. In terms of game design, the perceptions definitely fit as one of the aspects since it is the way that we interact with things outside of our body. The motor movement is also very important in game design as new game consoles like wii encourage users to have more movements than a traditional console where users only had to push bottons. The movement of hand and eye is also very important as they are the basis of what modern days games are about. For my group, we are making a flash based speed game which would require perception of the eye and the coordination of hand to be taken into account in our game designs.
Another interesting feature that might not apply to regular games but do indeed apply to educational games would be working memory versus long term memory. Since it is an educational game, we hope the players would intuitively learn the knowledge of what the game tried to teach through a series of games. We do not hope that they only have the knowledge in their working memory. Through continued playing of the game, we hope the users can move the educational information into long term memory.
Witton Chou 08:45, 6 October 2008 (UTC)
It is very important to consider the limitations of the human body when designing games. While the challenge of achieving a particular goal must have a certain degree of difficulty to make the game fun to play, as game designers, we must consider the aspects of the game that assist with playability as well as the barrier between what is difficult and what is impossible. If the player finds that a particular goal is way beyond their ability, he will become frustrated and disinclined to continue. However, if the goal is difficult but found to be within reach, this game balance will fuel the player's drive to continue. Similarly, if the design of a particular interface lends itself to better usability (such as the positioning and design of various indicators and buttons where information is easily extracted and useful tools easily found), then the user will have a better time utilizing the power of the program rather than having to decipher to layout. Some of the concepts in the article also lend themselves to the concept of using familiar designs as the recognition of these attributes is quicker, thus lowering the learning curve of the interface.
Cynthia T. Hsu 08:55, 6 October 2008 (UTC)
I think Fitt's law, on a conceptual (if not a detailed mathematical) level, is particularly relevant in games that involve tangible bits, or since those have not yet been commercialized, those games which require a joystick or a mouse. To a lesser extent, even games with arrow keys require the player to implement Fitt's laws; if you play Pacman, it's hard to resist the temptation to repeatedly press a particular arrow key, often leading to overshooting the particular turn that you had been aiming for. I suppose this is where feedback is important in a game/user interface; it helps with the correction that humans intend to do.
I found it interesting how they compared how physical matches, class matches, and choice reaction time were increasingly slower as compared to a simple reaction test because of the amount of cognitive processing involved. I think it's a particularly useful idea to keep in mind from the serious game perspective, since logically from a design principle standpoint, physical matches (e.g., keeping identical words to mean the same thing in the same area on each display screen of an interface) will make navigating a user interface much faster. However, in the domain of serious games, having the added obstacle of forcing the user to make a class match or a choice will increase the cognitive processing and help with encoding "long term chunks" and maintaining the game in the zone of proximal development. This relates to the Power Law of Practice as well.
I also thought Saliem brought up a very interesting point (which I overlooked) about how important it was to consider the processing time when producing feedback for a certain user response. In reference to my Pacman example, it's probably very very vital to include system delays in these calculations as well.
Frank Yang 08:58, 6 October 2008 (UTC)
The reading basically takes all human actions and quantifies them in a way one would a machine or a computer. When I came across the section about human memory, I immediatedly realized how important MHP could be in designing any UI. In games, any time spent learning the game can be attributed to the games "learning curve," but if the player is consistently forced to perform some unprompted action with some convoluted input, the game may be that much less enjoyable. Furthermore, location of alerts or status updates in a game is also incredibly important. If the player were forced to acknowledge every single alert due to domineering presence at the center of the screen, the flow of the game could be seriously disrupted. On the other hand, if important alerts were off to the side and easily ignored, the help it provides would be useless. Also, in certain games, executing certain "moves" with awkward combinations of joystick twirls and button presses may be frustrating if impeccable timing is needed. However, you cannot relax the timing on the combination too much in the off chance of it being input accidentally. The limitations of the human body must be considered in all aspects of designing any game, since humans are indeed the target audience for games.
Volodymyr Kalish 09:00, 6 October 2008 (UTC)
First of all, MHP is one of the quickest and cheapest ways to get a valuable and prompt feedback. The results of users interacting with MHP can be used to evaluate the interface of the serious game (if a user can work the interface with MHP, then the interface works for sure). Then, the serious game itself gets to be evaluated. Such an evaluation could show whether the game is worth investments or whether there should be a major redesign to make it more fun and interesting. Also, the results of testing the game via MHP can be used to determine the level of games difficulty and the steepness of the game's learning curve. All of these important things can be done way before the game is even attempted to be made using MHP and lo-fi prototyping.
Karen Tran 09:29, 6 October 2008 (UTC)
This reading was interesting, though difficult to follow at times. I liked reading the principles of operation. The Power Law of Practice, that the time to perform a task on the nth trial follows an inverse function, is something that makes sense; this is the first time I have seen it expressed mathematically with an equation. The Uncertainty Principle, that the time to make a decision increases with increasing uncertainty, also makes sense; it is interesting to see this principle expressed as an equation also. The reading mentions that access time is included in processor cycle time, yet does not explain why they are combined into one parameter. I think the reading should have provided an explanation of why these two factors are combined into one. It is possible that they are both included in the same parameter because the access time could be negligible or it could be that cycle time is of more interest than its two components separately. The motor system is what allows our body to move because it sends signals to our muscles and the muscles contract in the fashion that our brain designates it to contract. The cognitive system is the middle man that translates signals from the perceptual system to the motor system. If we could somehow use MHP to build our lo-fi prototype, then I think it would generate a really promising result. In serious games, we could use MHP to somehow estimate/anticipate the users’ reaction by measuring their performances and reactions. Although I’m not exactly sure to take into account all those math equations.
nathanyan 09:50, 6 October 2008 (UTC)
The reading was pretty interesting to me - I had never thought of breaking down the way humans operate (and interact with machines/computers) in such a concrete way. While we focused on the qualitative aspects, I found it really useful that the model was actually able to break down and derive equations - it helps to see exactly which factors come into play and to what degree that they're important.
For example, take the power law of practice. The idea that a task takes less time with practice is a common-sense notion; no equation needs to be derived to reach that conclusion. However, the power law of practice tells us that the time decreases at a rate of n^-alpha, giving us a way to actually quantify the time savings and thus weigh it against other factors.
For instance, with an alpha value of 0.4 (the reading says it usually varies from 0.2-0.6), here is the time scaling for each subsequent try:
Tries : Time
1 : 1.00
2 : 0.76
3 : 0.64
4 : 0.57
5 : 0.53
6 : 0.49
This information could be used in a design of a serious game, for example, to determine how many times a user should repeat a task. For example, if a game offered tutorial stages on a puzzle before finally letting the user play a full stage, it would be helpful to know the marginal benefit of making the user go through one additional tutorial stage - is the 0.53 to 0.49 time savings from going to 6 tries worth spending another X amount of time repeating the tutorial?
Trinhvo 11:01, 6 October 2008 (UTC)
MHP is very interesting in describing human perception in a very strange way. It's beneficial to thoroughly understand human perception to things to be able to design games with better user interface. In game design, two of the most important factors are visual and sound effects. If we understand how humans perceive images and sound we can get our users more attracted to our games. For example, the frame rate must be carefully calculated so it's not too fast or not too slow, otherwise it will hurt users' eyes and scare them away. In addition, if we can make sound effect better, users will act as if they are diving themselves into the game world. So they can react appropriately to this imaginary world. For example, in FPS, good sound effect can help players avoid enemies from behind or somehow locate where they are to help players survive in the game world. Another important factor in game design is human hand-eye coordination, so games with good control-handling will help players easily interact with the game's environment.
Juanpadilla 16:06, 6 October 2008 (UTC)
Interesting article! It did however remind of some 1950’s sci-fi movie. I think many aspects of this article can be brought over to game design, particularly Fits’ Law and the Power Law of practice. Other’s here have successfully argued that Fits’ Law does not relate to game play; however, I think it is important to consider it when looking at designing the game’s controller. The position of the buttons on the controller should be set to allow for quick access to those buttons used most frequently. Similarly, the same can be said for the Power Law of practice as many pc games use the keyboard for controls. Trying to minimize the keystroke combinations and possibly limiting them to single keystrokes with a more familiar keyboard position, might help with lowering the initial learning curve and keep some of those players who might have otherwise gave up on the game due to unwieldy control scenarios.
Yuta Morimoto 16:11, 6 October 2008 (UTC)
I think the concept of MHP is applied to evaluate the difficulty of a game. In reading, there are many examples about the time of task. Those examples told me that I can regard the human as computer only in the context of calculation the running time of certain task. This means that I can evaluate the difficulty of a task by measuring running time. Thus, we can compare various tasks in quantitative method. I think these aspects are able to be used to adjust the difficulty of a game. We can change the difficulty of a game easily by setting parameters. However, there is no obvious rule or criterion to define the hardness of a game. So, I think I can evaluate the difficulty of game by measuring the time of completion of task.
Anthony Kilman 16:41, 6 October 2008 (UTC)
The Model Human Processor (MHP) attempts to model humans as computers, with two primary components: processor & memory. Users fetch memories in response to external stimuli, and provide some form of a response. This model is very applicable to design of games in a general sense of HCI, in that we may treat the 5 senses as I/O devices (in a sense). In this manner, HCI can be modeled as an interface between two complex systems. By following this model, key constraints (such as a delays associated with actions in a program) are bound by some value, not only providing the programmer with a design template but additionally making the application itself much more engaging to the user.
Hao Luo 17:09, 6 October 2008 (UTC)
I don't see the Model Human Processor as a very useful way to help design games. What are we supposed to use this model for? I suppose we could use it when testing, to try to analyze how the user is playing the game with respect to his perceptual and motor systems. We could also ask the user questions about the game and see what he is aware of and what he is not aware of, which would involve analyzing his cognitive system. These are good things to keep in mind when testing your design, but they are also very general and we certainly don't need all the equations explaining exactly how fast the motor system responds to the perceptual system and so on.
On another level, I suppose we can consider the MHP in designing games rather than testing. We can consider how people will perceive games and how they will interact and react to games with their motor system. One aspect brought up in the readings is the arm-hand-fingers system. In a PC game, for example, the inputs are a keyboard and a mouse. On a console game, it's the controller. Two different methods of input that requires very different actions for the hands/fingers. We must then, modify our game and how it is perceived in order to accomodate for such differences. For example, shooters on PC are usually faster because our motor system is better adapted to a mouse and keyboard and we therefore have faster reactions when playing on a PC. These are of course, minor, but things to keep in mind when designing a game.
James Yeh 17:13, 6 October 2008 (UTC)
While the model human processor mostly focuses on low-level human interface interaction, it does seem to apply to games as well. In particular, the learning aspect of serious games is a foundation that should be built carefully by using information from the mhp; for example, the rate of transfer from working memory to long-term memory should be taken into account when game designers are incorporating their message into their game. Pieces of information should be chunked together to allow for higher retention, and not too much similar information should be provided at once, as to avoid the wiping out or loss of previously obtained information. In addition to memory writing time, many characteristics of the model human processor relate to the design of an effective game. For instance, reaction time and Fitt’s Law are two facets of the mhp that would clearly affect gameplay; objects that are farther away should have more tolerance for error when clicked, and no action should require a reaction time short than the fastman response. Moreover, the Power Law of Practice can be applied to games such that the material and gameplay later in the game should be novel and more difficult than the beginning material, so that the game doesn’t become to easy at the end as the result of practice.
Paul Im 17:19, 6 October 2008 (UTC)
MHP has been used in order to model the human mind as a computer. The mind receives inputs and produces outputs, and it also has different access times to retrieve information. This is very similar to my own field of study in Cognitive Science. The mind can be interpreted by a computational approach, much like these MHP studies.
There are, of course, many different relationships to the user interface of serious game design. A game can only be played if the player can keep up with what is going on; the game also becomes unplayable if it moves at too slow of a rate. In addition, the inputs need to be displayed correctly on the interface so that the appropriate outputs are produced by the player. The study of the user as a computer is very beneficial in order to create playable games.
Shyam Vijayakumar 17:29, 6 October 2008 (UTC)
Thinking of the human mind as a computer is an interesting idea and one that can be applied to serious games pretty well. Any game takes advantage of one of the five senses, mostly just touch and vision. These can be the I/O devices of the human computer. RAM could be the equivalent of a person's short term memory which kicks into action memory games. Hard disk could be equivalent to a person's long term memory. There are a lot of parallels surprisingly that suddenly come to mind when applying the computer model to humans. Ofcourse, this approach feels a little cold to me, so I would rather use the different approaches we've learned so far like contextual inquiry and heuristic evalution.
Jonathan Fong 17:50, 6 October 2008 (UTC)
As Shyam mentioned, this scientific/mathematical modeling of the human mind is lacking something that might be especially critical in developing UI's for games. Namely, the human aspect of emotion or spirit should be critical. Some of the calculations can be done to minimize mouse movement, maximize memorization, and have elements that are retained by the user. The bottom-line for the success of our games -- even though and especially because they are "serious" games -- is the "fun."
Mikeboulos 20:09, 6 October 2008 (UTC)
Breaking the Model Human Processor into three subsystems is a great way to construct our understanding of how human beings interact with computers, but most of all breaking them into interacting subsystems. namely: perceptual system, motor system, and cognitive systems. One of the components in these categories that I found most interesting was the motor system. we usually think of the perceptual and the cognitive systems when thinking of the software. I also learned that the human memory works with association, "The more associations the item has, the greater its probability of being retrieved". Keeping this in mind, when designing a serious game, we can try to reinforce our goal by trying to associate it with many things that might be in the long term memory as much as possible, so that it would be easier to retrieve as well as making it reside in the long term memory.
