In his quest to unlock the secrets of the human mind, Professor Albert Yeap has focused on two of its major faculties: space and language.
As Associate Head of the School of Computing and Mathematical Sciences at AUT University and director of the university's Centre for Artificial Intelligence Research, Yeap is in the right place, and his contagious enthusiasm guarantees a full research team.
"I want to know how humans become intelligent; in particular, how we build useful representations of large spaces in our minds, and how we acquire complex language at such a young age."
"While a psychologist would test theories of language and space by observing subjects, and a philosopher would test them by debate, as a cognitive scientist I test my theories by looking at what a computer can learn and achieve," says Yeap.
Yeap has been fascinated by our minds' ability to map large spaces since he began his PhD at the University of Essex three decades ago.
His theory is that humans and other creatures map large spaces by defining local spaces and then link them together to create a "cognitive map". To prove this idea, he is creating a new series of robots known as Albots, each behaving as a unique species with its own cognitive ability.
So far, Albot0, enabled with sonar, has managed to create a spatial map and find its way home using distance and direction information just like some lower animals. Next up are Albot1 (with laser sensors) and Albot2 (camera vision) and finally Albot3, which Yeap expects will be able to find its way around complex spaces like crowded streets.
By building various Albots, Yeap argues, they will tell us much about how humans and other species understand space.
Yeap's computational theory of language is equally fascinating.
Psycholinguists argue that children must be born with innate language abilities, because environment alone cannot teach them the complexities of language. But Yeap disagrees, arguing it is possible to learn a complex language "environmentally".
In the course of proving his language theory, in a project funded by a $1.7 million grant from the Foundation of Research, Science and Technology, Yeap has helped to create the "Impressions" software.
Impressions demonstrates that computers can process large amount of text intelligently. Launched this year as part of the Pingar search engine which, will be marketed later as part of Microsoft Sharepoint product, it allows the user to analyse documents turned up in database searches without the need to even open them. Users can request data about the placement and frequency of chosen search terms, and Impressions can even collate the relevant paragraphs of the document into a free-standing, grammatically correct, summary document on a given search term.
Not only has Yeap shown that complex language can be learnt environmentally, he has also created a product with enormous commercial potential.
Yeap has boundless enthusiasm for the sport of scientific discovery, so he will be unlocking not just these mysteries but as many more as possible. A robot that thinks for itself is on the agenda. Scary? Not at all, says Yeap.
Click here to find out more.