DECO1006 - UNDERSTANDING DESIGN AND COGNITION How designers think HMMM!! REFLECTIVE CONVERSATION Paul Murty Design Lab - Faculty of Architecture Design & Planning, University of Sydney . Australia April 2009
Jan 18, 2016
DECO1006 - UNDERSTANDING DESIGN AND COGNITION
How designers think
HMMM!!
REFLECTIVE
CONVERSATION
Paul MurtyDesign Lab - Faculty of Architecture Design & Planning,
University of Sydney . Australia April 2009
Part 0
Story so farStory so far
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Story so far
Introduction & Design and SocietyPhilosophy in DesignLanguage of DesignModels of design
AI in design / Knowledge-based design
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
AI in design / Knowledge-based design
How designers think
• Could be overlapping or different views• Feel free to ask questions or interject • Be prepared to speak, as thinking designers• This is about you as much as anybody else
Today - Concepts
1A Designing and Thinking 1B Models and paradigms 1C Ways of studying design thinking
How designers think
Tomorrow - Research
2A Process oriented research 2B Content oriented research 2C Future research directions
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Aims
o How Design differs from Science, Art activities
o Knowledge and processes involved in design
Methods of studying design thinking
Understanding
o Methods of studying design thinking
o Issues involved in design thinking research
o How (some – many) designers think
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab,Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Part 1
ConceptsConcepts
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
o There is a growing general interest in designing as a distinct and vital form of dynamic intelligence
1A Designing and Thinking
When we consider design as an activity let's call it designing
a distinct and vital form of dynamic intelligence
o Driving this is a view that designerly ways ofknowing, thinking and acting may be identified anddeveloped as a natural intelligence of design
o So, what are designerly ways and what is their role in the creative experience of designing?
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2099
1A Designing and Thinking
Discussion
The ways designers think ?
o What are your experiences of designing?
o How have earlier lectures influenced thoughts about design?o How have earlier lectures influenced thoughts about design?
o Have you observed the ways people design?
o How does your way of designing compare with others?
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2008
o Intentional: aims to change something
o Strategic: directed towards a representation, or prescription of a future end product or condition
o Communicated: to others, to act upon
1A Designing and Thinking
What do we know about designing?
o Communicated: to others, to act upon
o Situated - many actions and other factors may influence or assist the designing, or production of the end product
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
How does designing differ across disciplines?
1A Designing and Thinking
How is designing different from Science and Art ?
o Science - ing? - scientific methodo Art - ing? - creative expression
How does designing differ across disciplines?
o Architectural design? o Planning? o Product design?o Digital design?
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
o Challenge - major research frontier
o Understanding - of who and what we are
1A Designing and Thinking
Why study design thinking?
o Paradigm - applied intelligence in dynamic situations
o Raise design standards - improve quality of life
o Restore damaged world - a sustainable existence
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Long term Objectives
1A Designing and Thinking
Research Aims
o Explore and develop understanding of design thinking o Formalise knowledge in theories, models, methods
Long term Objectives
o Improve education of designerso Augment practice - develop tools to assist designerso Facilitate wider application of design
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
1B Paradigms and Models of Designing
Paradigm
o Model, or a clear and typical example of somethingo Conventional way of doing or thinking about something
Model
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Model
A representation of something
eg. a smaller physical object, or a simple description
Computational model
System having similar functions and relationship structure,
to the process it models
Broadbent (1973)
o Pragmatic – directly shape the materials of the artefact
o Iconic - adopt successful solutions as ideal forms (icons)
1B Paradigms and Models of Designing
Design Paradigmsor alternate conceptions of designing
o Iconic - adopt successful solutions as ideal forms (icons)
o Canonic – formalise rules (canons) from icons
o Analogical – use analogue medium, to represent design
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
1B Paradigms and Models of Designing
Paradigms - alternate conceptions of designing contd.
Gero (1990)
o Routine - proceeds from existing prototypes
o Innovative - proceeds from prototype, but with freedom to change (eg. stretch) some variables
o Creative – new variables, from a variety of origins
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
o Creative – new variables, from a variety of origins
Bi-polar alternatives
o Top down - conjecture an overall design first
o Bottom up - commence with design of parts
Top Down design
Iconic design
Treats successful solutions (prototypes) as:1.Templates for new structures2.Accepted form for structures of a type.
Routine design
Proceeds from existing prototypes
Canonic design
Builds upon iconic design by providing rules, or components, as design resources.
Innovative design
Proceeds from prototypes, with freedom tochange the ranges of prototype variables.
Analogical design Creative design
1B Paradigms and Models of Designing
Analogical design
Analogue medium, such as a drawing, is used to simulate or represent an invented design.
Creative design
New variables used, producing new types.Provides capacity to produce a paradigmshift
- Combination- Transformation- Analogy- Emergence- First Principles
Pragmatic design
Employ and manipulate materials directly until a suitable solution is created.
Bottom Up design
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Asimow (1962)
o Analysis – Synthesis - Evaluation
1B Paradigms and Models of Designing
Early models of designing
o were based on problem solving analogy o viewed design situations as problems
o Analysis – Synthesis - Evaluation
o Iterative: not a simple A - B - C process
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
ANALYSIS SYNTHESIS EVALUATION
o Problem-solving analogy has been criticised as being simplistic, negative, mechanical
o Many designers view a new project as an opportunity
o Requirements are often created during designing
1B Paradigms and Models of Designing
Design Problem view
The view of architect Denys Lasden. cited by Nigel Cross (Cross 1999) hints at something more:
“.. our job is to give the client … not what he wants, but what he
never dreamed he wanted ..”
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Designing and Problem solving are prompted by the presence of the unwanted or absence of the wanted.
o Design situations, or problems, are variously described as:
- ill-defined or ill-structured: “ill” meaning insufficient
- wicked: meaning complex and hierarchical, many interactions, eg. where you need to propose a solution in order to understandthe problem (Rittel 1984)
1B Paradigms and Models of Designing
Designing is problematic
the problem (Rittel 1984)
o Designing has also been described as rhetorical, exploratory, emergent, opportunistic, abductive, reflective, ambiguous, risky (Cross 1999)
o Necessary to clarify or identify the real problem(s)
o Uncertainties tend to require subjective interpretations
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
o The design and the designer are both situated
- The design has a context, the designer has an agenda
- The problem (barrier or unknown) always exists relative to the problem solver. Mayer (1989)
o The situated view recognises the dynamics of designing
1B Paradigms and Models of Designing
and: designing is situatedSituated = being in and a part of a dynamic context
o The situated view recognises the dynamics of designing
- Design is experienced as a situation which the designer inhabits. (Schon 1983)
- First person view - interactive, adaptive (Gero 2006)
o Design solutions
- Possibly no right solution, just better or worse (Rittel 1984)- Designs may consist of part solutions, or be holistic concept depending on many indeterminate and situational factors
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
o Do architects, other designers, artists, scientists, or people in other professions think differently in some systematic way? - In what ways might they differ?
o Which activities are the most designerly?
1B Paradigms and Models of Designing
Discussion
What do we understand by ‘design thinking’ ?
o Which activities are the most designerly?- Client meetings or site inspections?- Sketching, drawing, modelling? - Individual or group activity?
o Any comments on your designing and experiences of:- Different stages?- Different activities?- Your own style?
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
1C Ways of studying design thinking
Early developments
o The study of design thinking began almost before design as we know it today.
o Creative, expressive, analogical designing was a rarity before modernism became dominantbefore modernism became dominant
o The study of thinking was part of an evolving modernism
- Structuralism - focused on mental structures (W.Wundt 1880s)
- Functionalism - focused on what people do and why (W.James)
- Pragmatism - ‘knowledge validated by usefulness’ (J.Dewey)
- Associationism - how experiences, near in time, become associated and bring about learning. (H.Ebbinghaus, E.Guthrie)
- Behaviourism - proposed associations are related to satisfaction, not just contiguity. (E. Thorndike)
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Conflicting Views
In 1927 Philosopher Bertrand Russell wrote of a consistency in behavioural studies, of animals:
- The observed animals confirmed the philosophy believed by the observer
before his observations began
- Animals studied by Americans rush about frantically, with an incredible
display of hustle and pep, and at last achieve the desired result by chance
- Animals observed by Germans sit still and think, and at last evolve the
1C Ways of studying design thinking
- Animals observed by Germans sit still and think, and at last evolve the
solution out of their inner consciousness Russell (2007) For background see: http://forum.dcc.ac.uk/viewtopic.php?t=164
What makes this more than a funny story is that:
- Learning theory, the American perspective, developed from behaviourism dominated human psychology for several decades
- Gestalt theory, part of the German perspective and a vital part of the study of cognition became unfashionable until the 1950s
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
1C Ways of studying design thinking
Gestalt Theoryo Gestalt psychologists
- Focused on process and organisation
- Had a systematic perspective onwholes, having particular properties,that are not evident in the parts
- Rejected the Behaviourist view of thinking as simplistic
- Studied novel problems, requiring a qualitatively different approach
- Argued, solutions to such problems unlikely to arise by chance
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Mayer (1995)
o Max Wertheimer (1880-1943) proposed:
- Solving a non-routine problem requires insight into the problem,
or recognising how parts of a problem fit together, typically in a way not previously appreciated
- Grasping the internal structure of a problem situation is the central component of what he termed productive thinking, as distinct from normal reproductive thinking. Wertheimer (1959)
Gestalt theory focuses on process and organisation
o Preparation - when the problem is investigated or confronted
o Fixation – when barriers to solution are encountered
o Incubation - a mulling over stage, when the problem solver ceases to work on the problem, typically after getting stuck or experiencing some form of mental block
1C Ways of studying design thinking
Gestalt Theory contd
some form of mental block
o Restructuring - changes in awareness of the problem or goals, or or a reinterpretation, leading to insight
o Illumination - The sensation of welcome and gratifying surprise marked by the feeling, summed-up by, "Aha!"
Gestalt theory has had a long association with design perception but its influence in relation to design cognition is less apparent
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
A fresh approach to psychological research in the 1950s:
o Combined elements from different sources - Experimental research methodology - Recognition of learning and memory as keys to cognitive processes. - Combined concepts and techniques from German (Gestalt &Wurzburg) theorists, computer science, information theory and linguistics.
Rejected the Behaviourist view
1C Ways of studying design thinking
Cognitive Science
o Rejected the Behaviourist viewViewed humans as active, information-seeking and information- using organisms, rather than passive receivers of stimulation.
o Focused on information processing Viewed cognition as being fundamentally computational in nature - brain corresponds to hardware- mind corresponds to software
Refs: Reynolds & Flagg (1983), Mayer (1995), Simon (1999)
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Has developed at the same time as cognitive science
o Features of design thinking research - Vigorous field, developing quickly internationally
- Beyond “anything goes” - becoming more rigorous
- Draws on cognitive science and contributes to it
- Wide range of research options available
- Methods have different strengths and weaknesses
1C Ways of studying design thinking
Design thinking research
- Methods have different strengths and weaknesses
o Research Methods - Introspection
- Interviews
- Questionnaires
- Input-output experiments
- Protocol studies
- Brain studies
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Introspection – reflection
o Default method – research typically starts this way
o Long tradition of introspective research
o Quick, intuitively satisfying to author
1C Ways of studying design thinking
Design Research - Options
o Author reliability may be an issue
o Tends to generate un-testable conjectural theories
o Limited without other forms of validation
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Interview studies
o Direct means of acquiring first-hand information from source
o Strong indicative capability – unexpected detail, new findings
o Reliability issues: - Based on respondent memory - Questions may lead the respondent
1C Ways of studying design thinking
Design Research Options contd
- Questions may lead the respondent- Interviewer may bias results
o Reliability enhancement:- Quality and quantity of respondents- Semi structured, open ended questioning- Clearly specified coding, allowing replication- Multiple interviewers and coders
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Questionnaires
o Popular means of investigating large numbers
o Strong indicative capability, but limited follow up
o Reliability affected by being: - memory-based
Design Research Options contd
1C Ways of studying design thinking
- memory-based- biased by questions- biased by expectations
o Reliability enhancement: - Rigorous pilot study process- Large, objectively selected populations - Rigorous coding and data analysis- Statistical significance- Follow up interviews, focus groups or other research
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Input–output experiments
o Method Basic behavioural experiment: subject-task-action- Designer (subject) regarded as a black box- Change input (Brief), measure change in output (Product)
- May be carried out on humans or by computational methods
o Example :
Design Research Options contd
1C Ways of studying design thinking
o Example - study of fixation among designers, found: - Exposure to existing design influences current designing- Different forms of fixation affect different designers- Fixations occurred with both analogical (deep) and superficial (surface) features. (Purcell and Gero 1996)
o Applicability- Indicative of possible process (ie.fixation) causing a result- Doesn’t explore reasons or provide model of designing, only models the process that produces a behavioural change
- Computational models are strong proof of working processes
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
1C Ways of studying design thinking
Design Research Options contd
Protocol studies
o Procedure- Video+audio record designer while designing- Collect recorded data from:
. concurrent/think-aloud and/or
. retrospective verbal utterances- Convert utterances or events into codes- Analyse codes – graphical or statistical
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
- Analyse codes – graphical or statistical
o Research may be process-oriented and/or content-oriented - Think aloud reports better for process analysis- Retrospective more informative for content analysis
Protocol study of expert v novice designers
o Respondents are given a design task
o Analysis - distributions are identified by coding- event durations- micro strategies- drawing versus other activities
1C Ways of studying design thinking
Design Research Options contd
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
0%
25%
50%
75%
100%
0:00:00 0:10:00 0:20:00 0:30:00 0:40:00 0:50:00 1:00:00 1:10:00
% Function & Behaviour
Averaged Over 10 Minutes
Novice designer Expert designer
Protocol study of expert v novice designers contd.
o Analysis - key differences are identified
- Productivity
- Rate of cognitive activity
- Structure of concurrent actions
1C Ways of studying design thinking
Design Research Options contd
- Structure of concurrent actions
- Strategic knowledge
o Key finding
- Shows importance of strategic knowledge
- Experts work with bigger ‘chunks’ of knowledge
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Brain studies
o Fast developing research area
Technology exists to correlate brain activity with other behaviour
o Avenues of investigation include
1C Ways of studying design thinking
Design Research Options contd
- Neurophysiology - study of the
functioning of the nervous system- Brain scans . fMRI - magnetic resonance imaging. PET - positron emission tomography
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
How designers think
End of Part 1
Concepts
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
2A Process oriented research
How designers think
Part 2
Research
2A Process oriented research 2B Content oriented research 2C Future research directions
What follows next is a review of a variety of studies which all contribute, in different ways, to a better understanding of design thinking
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
2A Process oriented research
Examples
Computational research
A computational model of curiosity
Protocol studies
Solution driven versus problem driven design
To sketch or not to sketch?
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
2A Process oriented – computational research
A Computational Model of CuriositySaunders & Gero (2001)
OverviewStudy involved development of a “curious design agent”
Key ideas
o Curiosity - is the motivation to discover new knowledge when
faced with an unfamiliar situation. It can be used to guide the search and exploration of unfamiliar design spaces to find new knowledge and better understanding of a non-routine design task
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
knowledge and better understanding of a non-routine design task
o Curious processes
. Curious search can be used to guide problem solving
. Curious exploration can be used to guide problem finding
Novelty detection
The curious design agent uses a computational process called novelty detection to guide its search during the course of a design session.
Interestingness based on novelty depends upon the knowledge of the agent and its computational abilities; things are boring if either too much or too little is known about them
2A Process oriented – computational research
A Computational Model of Curiosity - Saunders & Gero
Experiment
To illustrate the behaviour of a curious design agent the researchers developed a computational model of a children's Spirograph design generator
The goal of this experiment was to examine the behaviour of a curious design agent as it searches the space of patterns that can be generated using the simulated Spirograph.
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
The results from the experiments suggest that providing design agents with a sense of curiosity confers significant advantages in the search of ill-defined design spaces.
2A Process oriented – protocol research
Solution driven v problem driven design
Kruger & Cross(2006)
Overview Data from protocol studies of nine experienced industrial designers, performing the same task, were analysed to develop an expertise model of the product design process
MethodProtocol statements and the expertise model were used to identify
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Protocol statements and the expertise model were used to identify four cognitive strategies employed by the designers:
- problem driven - High frequency data gathering and identifying constraints
- solution driven - High frequencies in generating and assembling solutions
- information driven – Very high frequency of data gathering, low frequency
solution generating.
- knowledge driven - High frequency modelling activity (utilising knowledge)
2A Process oriented – protocol research
Solution driven v problem driven design
The strategies and processes were then related to task outcomes, such as solution quality and creativity
FindingsThe different strategies were not related to overall solution quality in a simple, better or worse, way. Instead:
o Designers using a solution driven strategy tended to have
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
o Designers using a solution driven strategy tended to have lower overall solution quality scores, but higher creativity scores.
o Designers using a problem driven design strategy tended to produce the best results in terms of the balance of both overall solution quality and creativity.
2A Process oriented – protocol research
To sketch or not to sketch?
Bilda, Gero & Purcell (2006)
Overview Researchers conducted think-aloud experiments with expertarchitects to test whether sketching is essential for conceptual designing.
Procedure
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Procedure The experiment required the respondents to design in twoways, in two sessions;1. Respondents not allowed to sketch 2. Respondents allowed to sketch
The second sessions took place about 1 month after the first
The protocols were then analysed
It was expected that the two modes would require different processes
2A Process oriented – protocol research
To sketch or not to sketch? - Bilda (2006)
Findings
No significant difference between sketching and not sketchingbased on three criteria: 1. design outcome2. cognitive activity 3. idea links
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
3. idea links
ConclusionSketching is not an essential activity for expert architects in the early phases of conceptual designing
2B Content oriented research
Examples
� Interview studies
The primary generator
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
The primary generator Discovery processes in designing
� Protocol studies
Quantifying coherent thinking in design
2B Content oriented research – Interview study
The Primary Generator Darke (1978)
Overview- Research involves a series of insights about designing, emerging while Darke was reviewing some earlier interviews she had recorded with skilled architects
- Darke discovered that many architects were able to design more rapidly by initiating a relatively simple idea, she referred to as a primary generator, early in their design process.
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
primary generator, early in their design process.
Conclusion- Designing can be viewed as a process of variety reduction, where potential solutions are filtered by the designer's knowledge and capacity to structure the problem in solveable terms
- Conjectures of approximate solutions need to be proposed early, as many decisions cannot be taken unless a solution-in principle is known
2B Content oriented research – Interview study
Discovery Processes in DesigningMurty (2007)
o OverviewStudy of discovery experiences of 45 skilled architects and designers
o Findings
- Most respondents (near 80%) make insightful discoveries which assist their designing
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
- Discoveries are rarely “out of the blue”
- Designers adopt distinctive methods for achieving breakthroughs during conceptual designing and are insightful in different ways
- Discoveries after not working, ('cold' discoveries) are important
- Widespread evidence of latent designing (latent preparation)
2B Content oriented research – Interview study
Discovery Processes in Designing - Murty (2007)
Focusing strategies
� Orienting: Solution – Process - Wholistic
� Scoping : Generic - Specific
� Framing - Influence of values
Action Styles
� Progression mode – Fluctuate – Steady - Both
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
� Progression mode – Fluctuate – Steady - Both
� Incessancy – Always on – Need to switch off
� Reactivation – Most discontinue
2B Content oriented research – Interview study
Discovery Processes in Designing - Murty (2007)
Different types of discovery experiences
� Ideas – Solution/generator - “Aha!” discovery (19/45)
� Fluency - Succession or flow of ideas (5/45)
� Clarity – Revelation or awareness of new relations (10/45)
� Recognition - Item seen but not appreciated earlier (5/45)
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Apparently different levels of insight
� Incremental – Methodical or not clearly insightful (6/45)
� Insightful design - Hot discoveries, designing (14/45)
� Cold discovery - Not designing (15/45)
� Cold discoveries more insightful than hot (10/45)
2B Content oriented research – Interview study
Discovery Processes in Designing - Murty (2007)
Latent Preparation
� Like incubation – associated with discoveries
� Includes background activities - not entirely unconscious
- Ideas continuing to tick over
- Several problems percolating away at the same time
- Chewing over – letting things float around for a while
- Like friends - “chattering away, in my head”
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
- Like friends - “chattering away, in my head”
Conclusions
� Many different ways of being a good designer
� The natural intelligence of designing may be interpreted as a combination of many adapted individual attributes
� Becoming a good designer is a design problem
Quantifying coherent thinking in design A computational linguistics approach
Dong (2004)
o OverviewStudy evaluates a method for measuring coherence of communication in
design team conversation, based on latent semantic analysis
Patterns of interrelations between individual ideas and the group’s
2B Content oriented research – Semantic analysis study
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
Patterns of interrelations between individual ideas and the group’s ideas are also revealed by this analysis
o Key ideas- Design team conversations reveal thinking patterns and behaviour as
participants communicate their thoughts through verbal communication
- The study applies computational techniques that have been successfully applied to design communication in text, to conversational mode.
2B Content oriented research – Semantic analysis study
Quantifying coherent thinking in design - Dong (2004)
o MethodTranscripts of four engineering/product design teams communicating in a synchronous, conversational mode during a design session were analysed
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
o Conclusion
- Team’s verbal communication offers a fairly direct path to their thinking processes
- Link between coherent conversations and coherent thinking
2C Future research directions
Future Trends
Computational research
- Generally - closing in on human thought
- More sophisticated modeling of cognition
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
- More sophisticated modeling of cognition
- Computational modeling of brain functions
Human research
- Growing need for and utilisation of, design
- Greater diversity of design studies
- Greater need to integrate research
2C Future Research Directions – Computational Research
Literal the Blue Brain projectMarkram (2009)
OverviewSupercomputer (named Literal) with 8000 interlinked processes, each designed to replicate a real neuron in a real brain
Director of the project, neuroscientist Henry MarkramDescribes it as: “the first model of the brain... built from the bottom-up”.
“I wanted to model the brain because we didn't understand it," he says. "The best way to
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
“I wanted to model the brain because we didn't understand it," he says. "The best way to
figure out how something works is to try to build it from scratch... There are lots of models
out there, but this is the only one that is totally biologically accurate. We began with the most
basic facts about the brain and just worked from there”
The system accurately simulates a neocortical column, a tiny slice of brain containing approximately 10,000 neurons, with about 30 million synaptic connections between them.
"The column has been built and it runs... Now we just have to scale it up... In ten years, this
computer will be talking to us."
x
2C Future Research Directions – Human Research
Design research A revolution-waiting-to-happen - Dorst (2007)
o Overview - Dorst advocates a new wholistic design research, connecting the process and content of design activity with a modelof the designer and the context in which designing is taking place
o Key points- Design research remains ‘pre-scientific’.- Research needs to address design activity as a whole
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
- Research needs to address design activity as a whole. Design activity beyond the ‘design project’ overlooked
eg. higher-level activities, like work of senior designers
- Proposed methods need testing and informed guidance on:. Where, when and how they are applicable. The designers who will apply them. An explanatory framework or manual, of how to apply them
o Proposal– Deeper understanding of designing
based on consideration of all aspects of design activity
- A new kind of design research.
Questions
DECO1006 - UNDERSTANDING DESIGN AND COGNITIONDesign Lab . Faculty of Architecture Design & Planning . University of Sydney . Australia April 2009
References
Significant sources
Cross, N. (1999) Natural intelligence in design. In Design Studies 20, 1, 25-38.Darke, J. (1978). The primary generator and the design process. In W. E. Rogers and W. H. Ittleson(eds), New Directions in Environmental Design Research: proceedings of EDRA 9 (pp.325-337). Also in N. Cross (Ed.) (1984) Developments in Design Methodology, New York, John Wiley & Sons.
Lawson, B. R. (1997) Design in Mind. Architectural Press, Oxford.
Lawson, B. R. (1997) How Designers Think. Architectural Press, Oxford.Lawson, B. R. (1997) How Designers Think. Architectural Press, Oxford.
Rittel, H. W. J. (1984) Second-generation design methods. Horst W.J. Rittel, interviewed by Donald P.
Grant and Jean-Pierre Protzen. In Developments in Design Methodology, Cross, N. (Ed.) The Open
University. New York:John Wiley & Sons, 1984. Originally published in The DMG 5th Anniversary Report:
DMG Occasional Paper No. 1 (1972), pp.310.
Rowe, P.G. (1991) Design Thinking. Cambridge, Mass. MIT Press
Schon, D.A. (1983) The Reflective Practitioner. New York, Basic Books.
Simon, H. A. (1996) Sciences of the Artificial. Third Edition. MIT Press, Cambridge, Mass..
References contd
Other cited or useful sources
Bilda, Z, Gero, J.S., and Purcell, T. (2006) To sketch or not to sketch? That is the question. In Design
Studies 27, pp.587-613.
Bilda, Z, and Gero, J.S. (2007) The impact of working memory limitations on the design process
during conceptualization. In Design Studies 28, pp.343-367.
Broadbent, J. (1973) Design in Architecture; Architecture and the Human Sciences. John Wiley &
Sons, London.
Cross, N. (2004) Expertise in design: an overview. In Design Studies 25 (2004) 427–441Dong, A. (2004) Quantifying coherent thinking in design: A computational linguistics approach. In JS Gero (ed), Design Computing and Cognition'04, 521-540. Kluwer Academic Publishers, Dordrecht,Dorst, K. (2007) Design research: a revolution-waiting-to-happen. In Design Studies 29 (2008) 4-11Dorst, K. (2007) Design research: a revolution-waiting-to-happen. In Design Studies 29 (2008) 4-11
Dorst, K. and Dijkhuis, J. (1996) Comparing Paradigms for Describing Design Activity. In Cross, N.,
Christiaans, H. and Dorst, K. (Eds.) Analyzing Design Activity. John Wiley & Sons, New York
Duncker, K. (1945). On problem solving. Psychological Monographs, 58(5), whole no.270.
Gero, J. S. (1990) Design prototypes: A knowledge representation scheme for design. In AI Magazine
11(4): 26-36.
Gero, J. S. (1997). Concept formation in design: towards a loosely-wired brain model, in L. Candy and
K. Hori (eds), Strategic Knowledge and Concept Formation Workshop, Loughborough University of
Technology, Loughborough, pp.135-146.
Gero, J.S. (2006) Understanding Situated Design Computing and Constructive Memory: Newton, Mach,
Einstein and Quantum Mechanics. Unpublished?
Gero, J.S. & McNeill, T. (1997) An approach to the analysis of design protocols. Design Studies 19
(1998) 21-61
.
References contd
Goel, V. (1995) Sketches of thought. MIT Press, Cambridge, MA.
Goldschmidt, G. (1991) The dialectics of sketching. Creativity Research Journal, 4, 123 -143
Hillier, W., Musgrove, J., and O'Sullivan, P. (1972), Knowledge and design, in Mitchell, W. J. (ed.)
Environmental Design: Research and Practice, University of California.
Jansson, D. G. and Smith, S. M. (1991) Design fixation. Design Studies, 12,3-11
Koestler, A. (1976) The Act of Creation, London, Hutchinson.
Kohler, W. (1969). The task of Gestalt psychology. Princeton, NJ: Prince-ton University Press.
Kruger, C. & Cross, N. (2006) Solution driven versus problem driven design: strategies and outcomes.
In Design Studies 27, pp.527-548
Lloyd, P. & Scott, P. (1994) Discovering the design problem. Design Studies 15 (1994) 125-140.
Mayer, R. E. (1989) Human Nonadversary Problem Solving, in Guhooly, K.J. (Ed.) (1989) Human and
Machine Problem Solving, Plenum Press . New York
Mayer, R.E. (1995) The search for insight. In R.J. Sternberg & J.E. Davidson(Eds.), The Nature of Mayer, R.E. (1995) The search for insight. In R.J. Sternberg & J.E. Davidson(Eds.), The Nature of
Insight. MIT Press, Cambridge, Mass. 1995
Purcell, T. and Gero, J. S. (1996) Design and other types of fixation. DesignStudies 17, 363-383.
Reynolds, A. G and Flagg, P. W. (1983) Cognitive Psychology. Little, Brownand Company, Boston.
Russell, B. (2007) An outline of philosophy. readcountrybooks.com . This is a re-print of what looks
like the original version. There are many earlier printings dating back to 1927. Simon, H. A. (1999). Karl Duncker and cognitive science. From Past to Future: The Drama of Karl Duncker, Vol.1(2), pp.1-11. Frances L. Hiatt School of Psychology, Clark University.Saunders, R. and Gero, J. S. (2001) A curious design agent: A computational model of novelty-seeking behaviour in design, in Proceedings of of the Sixth Conference on Computer Aided Architectural Design Research in Asia (CAADRIA 2001), Sydney, pp. 345–350.
Sternberg, R.J. (1995) In search of the human mind. Harcourt Brace, Fort Worth.
Wertheimer, M. (1959). Productive thinking. New York: Harper & Row.
Images
Slide15 - Asimow Model - Gero, J.S. How Designers Think3 - lecture
Slide22 - Gestalt images - Bilda, Z. How Designers Think1 - lecture
Slide32 - Novice and Expert Designer images - Bilda, Z. How Designers Think1 - lecture
Slide34 - Brain images - Bilda, Z. How Designers Think1 - lecture
Slide39 - Novelty detection image - Bilda, Z. How Designers Think1 - lecture
References contd