New Jersey Institute of Technology (NJIT)
Computer and Information Science Department (CIS)
CIS677:
Information System Principles
Professor: Michael
Bieber
V. Balasubramanian, E-Papyrus Inc., (bala@e-papyrus.com; http://www.e-papyrus.com/personal/balaindx.html)
Davis: General Model Of Humans As Information Processors (see handouts)
Davis: Filtering Information to reduce processing requirements (see handouts)
Newell-Simon: Human Information Processing System
(refers to the paper George A. Miller, "The Magic Number Seven, Plus or Minus Two: Some Limits on our Capacity for Processing Information", Psychological Review, Vol. 63, No. 2, 1956)
Example: Which is easier: Chunking a phone number 1-800-CISNJIT or 1- 800- 247-6548
(The point is that you've broken the number down into 4 chunks (1, 800, CIS, NJIT) instead of 9 chunks.
Newell-Simon: General Model of Human Problem Solving: Memory
Bob S. asks regarding 1-800-CISNJIT (11:28):
My question was related to your comment about the picture. Let's suppose you have the phone number for CIS NJIT and instead of putting the word CIS, if you put the picture of building, people would remember the number faster than if you put the word CIS.
Bob S. continues (12:12):
Right, the letters are part of the numbers but people think they are not.
During this segment, part of the joke got lost. We made the analogy of short-term memory, long-term memory and external memory and their computer equivalents. Then I asked what the equivalent of the computer technologies: Local Area Networks and the Internet would be - would these equate to ESP in people. Bala's reply referred to this, when he joked "I guess this is like plugging the network into your head". Then he moved on to discuss these as equivalent to Collective Memory.
Mark S. asks regarding organizational memory (16:48):
The relationship between organizational memory and a document management system, are they one in the same, or separate?
{note: this comes from the discussion on document management systems in lecture 5}
Mark S. continues (17:50):
Between turnover and downsizing and increased use of consultants, organizational memory becomes more important. It seems like there ought to be a lot more emphasis on that area.
(from the Newell-Simon General Model of Human Problem Solving: Task Environment)
The task environment is the problem as it really exists. The problem space is how a particular decision-maker perceives the task in order to work on it. In other words, the problem space is the decision maker's mental model of the task environment.
Regarding models and mental models being a perception , Marium notes (20:12):
There's also a comment on this in one of the articles, where humans process information, they are making mental models , and it was very interesting because it said that a bad model can be rational in the mental world but not necessary to be rational in the real solution. How do you account for something like that?
* in relation to Alter's Work-Centered Analysis
{slide/handout: Information Characteristics}
example: "I after E except after C" rule of thumb when spelling.
e.g., elimination by aspects:
Regarding heuristics, Bob S. asks (29:14):
I have a comment here. Do you consider all the legal and ethical issues rules of thumb also? Because there are the rules no matter what you do you have to follow so....Generally can you say the legal business is legal or ethical? Unless you follow some rules, guiding principles....
Mark Senegal answers (29:42):
I think it will limit your problem space...you know...when you take these things in the consideration.
Bob S. then continues (29:46)
Yes, but then another thing will be rule of thumb.
Bob S. later asks regarding supporting heuristics and guidelines (30:20)
You don't know. So then how do you handle the overload design of optimal information systems because it's a big overload by implementing all these rules.
Kevin M. asks (31:42)
Is there any typical heuristics that are going to be implementing the operations that the decision maker is using, in order to make decisions? So you might want to provide more criteria to base decisions on.
Example:
* filtering
Regarding which aspects to filter out, Mark S. notes (35:40):
That may be not a good assumption if you are in a rocky situation financially, you may do well to study some of the car companies not around anymore.
* Humans as Intuitive Statisticians
People believe they are better statisticians than they actually
are.
* Belief in the Law of Small Numbers:
People believe they can generalize from a small number of instances
(or in experimental terms: drawing conclusions from small sample
sizes).
- Availability Bias/Recency Bias
People only use information that is easy to recall (e.g., frequently
seen information)
- Concreteness
People use only information readily available *and* only information
in form needed
Anchoring & Adjustment
An anchor is base starting point, base for evaluating information and
for adjusting judgment as additional information arrives (people are
bad at measuring this base rate)
Mark S. asks regarding anchoring (41:15):
Has there been any research into the quality of anchoring? Because I know that there is a lot of going in wrong position. Sometimes you just grab it out of air but this is a place to start and from there you would adjust and usually end up with something that's not at all like what you started with. But a lot of time is wasted in that generation of all these adjustments, more from this optimal idea to what reality is.
Mark S. then asks regarding the starting point for anchoring (42:42):
If there are some guidelines about the way to find a good anchor then we can save some time.
- Representativeness Bias
How likely is it that an event/object/symptom being observed is a
certain type of event/object/problem?
- Dependence Bias:
People tend to believe that totally independent events are
related.
Regarding superstitions, Mark S. notes (45:22):
... superstitions are heuristics ...
- Cognitive Style:
Different people interpret information in different ways, and
different people tackle problem-solving in different ways.
Questions:
Do you think that IS designs take these biases and limitations into account?
Do you think that designers understand human biases and limitations?
How much does the field of HCI (Human-Computer Interaction) take these into account?
Why aren't most developers aware of these limitations and biases, and take them into account in their progams?
Susan C. notes regarding people's psychology (51:39)
It seems to me it's psychology. It's another field you have to understand: both information systems and psychology and put these two together.
More Questions:
Why isn't this intuitive for technical people?
Is it more difficult to design systems if you take cognitive limitations and biases into account?
Where is the true expense?
How much work is it for designers and developers to ensure that their systems work with people concerning their bias and limitations?
Do designers need to design new features to help people overcome biases and limitations?
Is it understood how you train people around cognitive biases and limitations?
So what would you recommend to managers of development projects?
This page: http://www.cis.njit.edu/~bieber/CIS677/lecture-notes/lecture8.html