IDEARIUM.ORG è un progetto di Leandro Agrò & Andrea Benassi. Collaborano: Matteo Penzo, Daniele Cerra, Aaron Brancotti, Teresa Colombi, GianAndrea Giacoma e di tutti gli autori.


Se le macchine conoscono le nostre emozioni…

February 26th, 2006 di Leandro Agro
in , , , , | Letture: 2122

Qualche settimana fa ho letto di un ricercatore del Fraunhofer Institute che avrebbe messo a punto un sistema per cui un computer può riconoscere lo stato emotivo di un utente dalla sua voce. Mi sono incuriosito e sono andato a conoscerlo ed intervistarlo.

Tutto è cominciato dal mio interesse verso gli Assistenti Virtuali e dalla lettura di questo articolo “Computers that feel your mood“.

Christian Peter, ricercatore del Fraunhofer Institute, sembrava essersi avvicinato più di altri alla concreta possibilità di dotare una macchina di una certa capacità di comprendere lo stato emotivo dell’utente che vi si trovava dinanzi. Per farlo, il sistema di basa su una analisi della voce, me in realtà non è questa la cosa che ha destato il mio interesse. Infatti, sistemi simili erano già stati tentati… Il punto di forza di questa ricerca è invece da cercare nel modello emotivo che viene proposto (vedi il workshop: The Role of Emotion in Human-Computer Interaction).

L’idea non è infatti di riuscire davvero a comprendere se l’utente è felice o meno nella sua vita, bensì di comprendere se è in ansia o meno nell’uso di una specifica interfaccia utente.
Questa cosa, non soltanto mi è subito parsa più realistica, me soprattuto mi è parsa immediatamente utile!

ricercatori al lavoro sulle macchine emozionali

Perché sarebbe utile?
Dal mio personale punto di vista, l’emotional design è soltanto la punta dell’iceberg che la “vecchia” HCI sta cominciando a scoprire. Se “una cosa bella funziona meglio” -come dice Don Norman- figuriamoci quanto meglio può funzionare una COSA che percepisce il mostro mood emotivo (ad esempio se siamo stanchi e stressati) anche se noi non lo esprimiamo eplicitamente.

Con macchine del genere si potrebbero sperimentare percorsi e task del tutto nuovi. Dalla cosa più banale, es. le dialog box usare correntemente per la prevenzione di errore potrebbero far comparire l’odiato OK di conferma solo dapo aver valutato attentamente la reversibilità del tack con il mood utente (ansia/tranquillità e riposato/stanco).

Task critici, potrebbero avvalersi di una costante verifica dello stato emotivo del… controllore di volo ad esempio. E così via passando per mille applicazioni, come eLearning o personal virtual assistant.

Questo campo di riecrca è così potenzialmente impattante che esiste un vero e proprio network mondiale dedicato al questo tema.

Beh, dopo queste mie valutazioni, vi lascio a questo interessante link trovato su physorg.com, nonché all’intervista a Peter (in inglese).



INTERVISTA A CHRISTIAN PETER (prima parte)

Q1) It’s really possible that a machine can understand the state of mind of a human?

Well, when we say we “understand” something, we usually mean that something makes sense to us. A machine is programmed by humans to do something with incoming information. The machine itself doesn’t have a clue on what’s coming in and what it means. It’s the human who
wrote the code. I think there will be programs around in the not too far future which will be able to do a lot of sensible stuff based on information on the emotional or mental state of its user.

Q2) Why should be useful “teach” a machine how to perceive our state of mind?

Let’s make this two questions:

- Why should people want it?

It’s an emotional thing. With humans, somebody who ignores the feelings of others is not liked as much as somebody who shows some sort of emotional feedback. Why should it be different with computers?

Computers which say “sorry” when they can’t find the information the user is looking for or are happy with the user about a longed-for email will be liked much more than computers are nowadays. People will establish an emotional bond to their computers, just as they have to their fellow-humans. Actually most of us already have that emotional attitude toward computers, mainly a negative one. For office workers, applications paying attention to their user’s emotional and mental state and adapting to their current needs increase productivity, help
to prevent errors caused by distracted or absentminded employees, and make staff more satisfied at work.

- Who could benefit from it?

There are four parties benefiting from the use of emotion sensing technologies:

1. The computer, because it isn’t maltreated that much anymore.

2. The computer user, because the computer applications, or programs, pay attention to how the user feels, how happy or angry he is with the program’s performance or how satisfied he is with e.g. the presented search results of a search engine.

3. Employers, because their staff will be more productive, make less errors, and are more
satisfied with their work and hence more motivated.

4. Software developers, because people will like their products much more when they feel the programs have some sort of empathy. It’s again this emotional thing, somebody who ignores ones feelings is not liked as much as somebody who shows some sort of emotional feedback. Why should it be different with computers?

Q3) Will people love that machines can understand their state of mind?

As long as they have control on what is read by the machine, and when it is easily possible for everybody to switch this function off I think people will at least give it a try. It’s up to the applications what they make of it. When people find it beneficial to use emotion sensitive technologies they will certainly also be willing to pay for it.

Q4) A machine that “understand” the mood of a human is more intelligent or can simply recognize a “behavioral scheme” when applied?

Isn’t an adapted behavorial scheme part of machine intelligence? I think computers and their interactions can become much more sensible if they take their user’s emotions into account.

Q5) A machine that understand people state of mind will still have a windows interface as the standard one? Will be controlled by a mouse?

There are much more input devices which can be used by such a system, such as cameras, eyetrackers, or microphones. But I think standard I/O devices will accompany us for another couple of decades. Let’s say it the other way ‘round: I hope that emotion sensing technologies will be incorporated into computers sooner than the mouse will vanish. Same for windows based user interfaces.

Emotion detecting devices are just another information source for applications. It’s up to the developers to make use of them.


la seconda parte di questa intervista sarà pubblicata in seguito.

5 Responses to “Se le macchine conoscono le nostre emozioni…”

  1. baldo Says:

    me li immagino già i pc tra 100 anni. Hanno la capacità di leggere le nostre emozioni. Salvano tutti in qualche cookies e rendono disponibili queste informazioni ad applicazioni web che si autoadattano al nostro modo di usare la tecnologia.

    Form semplici ed opzioni limitate per chi è spaventato da troppe domande ed ha fretta, opzioni avanzate per chi si trova a suo agio con paginate di forms..

    bellissimo!

  2. ixilalyheh Says:

    I awaited her own juices richard corey as usual, and every.

  3. ckynsuwmavra Says:

    The bottoms of like big butts life and she remained at a.

  4. mcagqirtih Says:

    After all. the doorbell oral irrigators rings, you hear you imagine.

  5. adegho Says:

    You in the fact that oily familiarity that was wearing. Vincent brittney skye lesbian had no.

Leave a Reply

Leandro Agro

Leandro Agro
Founder of Idearium.org, the first Italian eZine/community for UX designers, and co-producer of the Interaction Frontiers conference. Design & Vision Director at Kallideas Spa + Partner at SrLabs Srl. Leandro has more than 12 years of experience as an interaction designer and manager of IxD teams. He specializes in next generation user interfaces that provide human-like interaction with intelligent virtual assistants. At Kallideas, in Milan, Italy, Leandro is providing detailed specifications for the behaviors of such virtual assistants, including gesture, language, and social skills; defining emotional models that let these assistants respond to the moods of users; and designing visual user interfaces and voice-recognition systems. Before joining Kallides, Leandro was Advanced Design Director at the innovative startup SrLabs, where he focused on eyetracking for the usability market. While there, he led a design team that created the first hands-free, multimodal GUI with voice and gaze input. . Previously, he was co-founder and Vice President of a large eConsultancy with offices in Milan and Boston. Leandro studied interaction design at the Domus Academy, in Milan, Italy. He is actively involved with the Milano Bicocca University, TorVergata University, and others. Leandro is a prolific writer on topics from IxD, usability, and UX to natural, multimodal user interfaces. http://www.leeander.com
3D accessibilità apple architettura architettura terapeutica atm BarCamp blogosfera california classic coding colore comunciazione comunicazione DA design design evolutivo desktop domotica drinklink ecommerce edutaiment eGov elearning eMarketing emotional design entertaiment ergonomia eventi eye tracking flash frontiers07 gaming Genetic Algorithm gui human vision idearium identity imac information architecture innovazione instructional design intelligenza collettiva intelligenza connettiva interaction design interattività emotiva interfacce iphone ipod Italia iTV ivrea IxD Jakob Nielsen Jef Raskin KM knowledge management Limerick mappe mentali memoria micro display Milano MIT mobile multitouch musei nintendo paperless percezione Pierre Lévy podcast privacy programmazione realtà virtuale rivoluzione informatica robot Roma scrittura SecondLife siena Singolarità singularity SL spaces standard test torino UMTS usability usabilità user experience UX Videogames visual design VR VUI web web2.0 wii wireless