Psy/Orf 322

Human – Machine Interaction

Spring 2004


Helping Humans Make Better Decisions



Operations Research and formal Decision Sciences grew out of military and industrial needs to create a discipline focused on making better decisions about the challenges of

·       troop deployments,

·       re-supply strategies,

·       production processes and

·       organizational priorities, to name a few. 


The discipline has created computer-based tools that yield decisions about what to do, when! 


Some have suggested that individuals in daily life face similar, while not as daunting, decisions and could use similar tools.


Is this over-kill? 


Some have suggested that individuals in daily life face similar, while not as daunting, decisions and could use similar tools.




What do we do during the day:

1.      Sleep (8hrs)

2.      Work/Learning (8hrs)

3.      Eat (2 hrs)

4.      Commute (1hr +)

5.      Run errands (1hr+)

6.      Recreate/Entertainment/Learning (4 hr- + weekends)


What’s important:

1.   Food

2.   Shelter

3.   Clothing

4.   Travel

5.   Entertainment/Learning


Examples of every-day decisions for which individuals could use some help?



For example, the task of going to a particular place requires decisions by the individual of: where to go, when to leave, how to go, what route to take, turning left at the next intersection, etc.  Not surprisingly, consumer-oriented adaptations of military and industrial systems that aid personal decision making in route choice and personal finance are beginning to appear on the market. 


Is there a real need here, or are these simply solutions looking for a problem? 


Homework reported:




I am attaching two articles on the potential and already implemented uses of real-time optimization software to assist doctors with patient care and particularly with drug prescriptions.  I chose two articles because both were ~5 pages and contained different relevant information on the subject.


Amit Koren


Assignment 5

April 12, 2004


Paper: Movies2g0 A new approach to online movie recommendation

Rajatish Mukherjee, Partha Sarathi Dutta, and Sandip Sen


This paper describes a new and innovative movie recommendation system.  The fundamental difference between this system and other existing systems is the algorithm used to make the recommendations.  Most other systems employ nearest neighbor algorithms and utilize social filtering of movies (i.e., since you liked movies A and B, you will probably like movie C, because other viewers who liked movies A and B also liked movie C).  This system uses a voting algorithm, which utilizes user preferences about movie genre, actors, directors, and various other features in order to match and rank possible movies.  This system is obviously useful to people in making everyday decisions about which movie to go see or rent.


Attached is my homework #5 brief of an automated parallel parking feature that Toyota recently developed.


Cory Jerch


Vippy Wong

Source: Loadholtes, Nick. “Demystifying AI in Everyday Life.” Artificial Intelligence Warehouse, AI depot. Online. Internet. April 13, 2004. Available:


Artificial Intelligence has created machines that mimic human behavior and intelligence. Humans created AI systems by programming human behavior and knowledge into a language that machines would understand. The thought process and decision-making behavior of AI systems is based on human behavior; thus, the decisions made by these machines are similar to those made by humans. The article states that it is easier to understand decision-making skills from the perspective of an AI system. By studying how these machines ‘think’ and make decisions, we can better understand and possibly improve the decision-making behavior of humans, and therefore is pertinent to helping humans make better everyday decisions.



Lizzy Louis


Homework 5

13 April 2004


Helping Humans Make Better Everyday Decisions


Ian Olgeirson. “High-tech gift ideas for the home/For the rec room ... The Yardage Pro 800.”  The Denver Business Journal.  4 December 1998.  Online.  


The article cited above describes a golf gadget, the Yardage Pro 800, which determines the range to any particular object.  The golfer uses it to find the distance to the hole or across a bunker, which in turn, determines the club he or she will use.  The Yardage Pro 800 takes the uncertainty away from deciding how far away the hole is and consequently, deciding which club to use.  This helps human golfers by freeing up some of their concentration that would otherwise be used to calculate range and to think about which club to use, thus allowing them to focus more on their swing. 




Kate Barber

ORF 322

Homework 5


Title: “An Inquiry into the Nature and Causes of the Wealth of Kitchens”

Author: Phil Salin

Market Process, Spring 1990


        This article focuses on the theory underlying the most basic human decision making processes (using as an example the sequence of decisions made by a human while cooking a meal).  Salin posits that the fundamental, most basic human decisions are made via “tools” that have invisible functions implicit in their designs.  The article is pertinent to the issue of human decision making because it implies that basic human decisions are not easily replicated (since the tools used in making decisions do not come from fuctions which can be easily understood, much less duplicated).  






See attached.  The article describes the use of lie-detection equipment to assist people in making the common decision as to whether to believe someone else, the sort of decision that gets made in every social interaction we have.  People generally use intuition and a rather unreliable assesment of subtle stress factors to decide; these machines are capable of much more precise analysis.  Apparently there are a number  of difficulties associated with using these machines to detect falsity, however, as listed in the paper. 


-Austin Akey




Sophia Kim

Psy/Orf 322

Assignment #5

April 14, 2004



Vehicles may one day use sensors to avoid hitting objects



This article discusses the benefits of a parking sensor, an on-board computer which helps the driver in detecting the size and location of objects near the vehicle.  This is performed through a radar system that has been mounted on the bumper which sends out pulses.  The parking sensors currently in use warn the driver with beeps or lights on the dashboard.  This would aid people in making better everyday decisions when it comes to commuting and parking the vehicle.  It would help the driver detect any objects that may result in a collision by searching for and notifying the driver of any objects in the way.  He or she can then stop the car and either wait for the object to get out of the way or choose a different route.        


Notes on Consensus Decision-Making; Randy Schutt;


This paper explores issues associated with group-based decision making.

Different approaches to cooperative decision making are explored and paper

identifies various types of consensus. Practical application of this topic

is portrayed on an example of a group trying to decide which restaurant to

go to. The decision flow is analyzed and different aspects that lead to

consensus are described thoroughly. Several tips on reaching the consensus

in the light of each individual's decision making process are also offered.


Peter Fabian


It's a little longer than 10 pages, but very interesting in my opinion.

  Evan Haas



"Cognitive Technologies: The Design of Joint Human-Machine Cognitive Systems"
David D. Woods
The AI Magazine
Volume 6, Issue 4, Winter 1986, pages 86-92

This reading is pertinent to "Helping humans make better everyday decision" because:

This reading discusses the potential for artificial cognitive systems to either make decisions for humans based on a series of inputs by the human (data) or to offer a set of solutions and let the user decide which solution/decision is most appropriate.   There are many ways to tackle the problem of joint cognitive systems, and this paper discusses several of the different ways one could design a computer to aid in decision making, and the potential upsides as well as pitfalls associated with each.  How we decide to build an automated decision helper depends on how we define "helper."  If by helper we mean something that is called in to solve a problem, then we need to build very good decision makers, and have people call on these machines with problems.  If, on the other hand, we define a helper as something that acts as a source of information, then the human problem solver is in charge of solving the problem, but may use the machine to do probability calculations, but the human must ultimately select what to do.