‘All Have Similar form yet none is the same as the other, and thus the choir shows a secret law, a sacred mystery.’

– Goethe

“I am interested in the internal landscape of the human mind, the structure of emotions, thought and the nature of being. The relationship between the internal psyche of the individual and the external world; how these forces shape the individual and the resultant states of being. These explorations form the content of my work and take shape as drawings, object, video and audio installations.”, Anjana Kothamachu

The word ‘pareidolia’ translates to making meaning from random stimulus. Integrating a variety of media (drawing, animation, open source code and installation), Anjana Kothamachu’s work addresses the intersection, or divergence of creation and consciousness; whilst playing with the idea that we project our reality onto the world around us, rather than passively experiencing it.

In Anjana’s video, there is a construction and de-construction of the experience of reality whilst under duress. This video depicts the state of trauma as a movement from emptiness and alienation to bewitchment by dark powers which, in turn, leads to an eruption of huge affect. There’s a slow transformation of the protagonist that involves a breakdown in memory, awareness, identity and/or perception.

The narrative is an excavation and re-working of memory, bleaching and re- ordering reality in the process.

Pareidolia is made using drawing, digital tools and artificial intelligence algorithms.

The video originates from one still drawing (ink on paper) that appears at the beginning. Anjana has then modified the image using traditional animation techniques and digital tools like Photoshop, After Effects and Nik Collection. These images were then run through an image recognition program. The program uses an algorithmic feedback loop from the artificial neural network. The software was originally intended for the ImageNet Large-Scale Visual Recognition Challenge(ILSVRC). This program is designed and trained to detect faces and other patterns in images, to automatically classify them.

When the program is fed with an image (in this case my drawing), it attempts to recognize patterns within it that might not be obviously apparent. The algorithm can also modify the image slightly to reinforce and enhance the patterns that  the neural network “sees”. The software repeats this modification in a loop till the nature of the patterns becomes apparent.

The process is akin to the human tendency to perceive patterns even when looking at apparently vague or obscure stimuli like clouds or ink blots. The mind tends to reinforce the patterns it initially notices.

At each stage of interaction between the program and my image I have modified the basic parameters of the program guiding it to respond in particular ways. Following numerous repetitions of the above process, I was able to generate multiple layers of enmeshed imagery. These images where then again manipulated and fused into one fluid series to create the final animation.

The final installation features multiple complementary actions – the animation screened onto a manufactured surface; another motorized mirror onto which images are projected. This in turn generates moving reflections that are projected onto the walls of the room. In addition, strategically placed mirror pieces create other still and moving reflections. This is accompanied by a voice over telling a story.